I'm pretty sure FB amplifies the posts of people who it wants to entice to use it more. It's certainly seemed to for me when I dropped back in after my usage falling off. Without data, it's just as easy to guess that people don't feel like interacting with your posts as The Gods Of Social Have Blighted You
Currently, it is relatively easy to mirror Facebook into an RSS feed + static website. I even had a pet project doing exactly that some time ago. If you really want, you can even mirror your friend's posts. Might be easier, than to convince people to move.
We've got something like Roam Research (except geared towards narrative rather than academics) that we want to explore adding a social network and feed capabilities to.
I'm pretty sure it's a pipe dream to have that feature in a social network that matters. We're pretty well past the era where non-commercial social networks could grow much past the curiosity stage, and commercial social networks need a way to monetize themselves. Since asking users to pay to join a social network is a non-starter that leaves ads as the monetization mechanism. It's easy to build what you describe from a tech standpoint, very hard to do so from a social network standpoint.
I would like that and also forum comments that are ranked by what I would find agreeable instead of what some anonymous, amorphous, majority of the moment finds agreeable. Why should the user always be made to feel like the majority view is the "best" or "most accurate" view of things?
At the core, I would do this by building a network graph of users who I've upvoted and users they've upvoted, and so on maybe a few levels deep. If anyone in that group of users has upvoted a comment, I want it on top - sorted by how many degrees of separation there are between me and the upvoter.
The very simplistic, single-faceted comment rankings that most places implement right now is, including Hacker News, is simply brain-dead IMO. I've brought this up before and 100% of the time, one of the responses I get is "sounds like a good way to build an echo chamber". It's not a bad thing actually. I think it more accurately models real life where we don't hang around with a bunch of people that we find disagreeable, even if they are in the majority and even if they're "right". With a system like the one I propose, everybody can feel comfortable, say what they want and things will most likely be way, way easier to moderate if they don't just moderate themselves.
The important part isn't the feed, but the business model.
If your users' needs are important to you, then your users need to be your customers - i.e. they need to be paying you.
If someone else is your customer, then their opinions and needs become more important than your users' opinions and needs.
The problem with Facebook is that advertisers are their customers and therefore they're the people that Facebook needs to keep happy. Providing a great experience for their users is secondary.
As users we get annoyed by this, but we're stuck with it until Facebook starts giving us the option to pay for our FB accounts. Which Zuck promised he'd never do.
So for your new shiny social network, the important question, as always, is: who's going to pay for it?
I think the flaw in your question is the assumption that a social network needs a business model. It's like asking what the business model is for SMTP. It shouldn't be under the control of a single entity so it shouldn't need a single specific business model. Then you have a market for providers even though it's still all the same network.
The business model of SMTP is that e-mail providers leverage their compliance with SMTP to draw users in but then derive value some other way (either by luring users into their ad-powered ecosystem like most consumer e-mail providers or making them pay like business e-mail providers).
A protocol by itself indeed doesn't need a business model but clients for the protocol do need one.
> A protocol by itself indeed doesn't need a business model but clients for the protocol do need one.
They still don't need one, there can be many. Even if most people still end up with the gmail model, it's meaningfully different, because the switching cost is lower. You don't have to convince everybody you know to switch too.
Now suppose you're trying to solve the original problem. You've got feeds from a hundred users and you want a way to filter them yourself. If any service exists that provides this, using any business model, you can use it even if hardly anybody else does. You can also develop your own client that does this or commission someone to add the feature to an existing client.
More than that, if this is a feature many people want, now the ad-supported providers will need it too, because now they have to compete for users with the providers who are treating you as the customer instead of being insulated from competition by the network effect when they own the entire network.
True. You could create an open standard for this, and then people can self-host their own social servers. I don't think that's what the gp was proposing, though.
And that's not going to be the norm. Most people don't want to self-host their email server, or even pay for it. They're happy trading their privacy for free gmail, and most don't even realise that this is a problem.
So as Nextgrid pointed out: that shifts the problem one step down the ladder. A commercial social client requires someone to pay for it.
Feeds in social media are severely limited for when you need to filter the content you want. What if you let people program their own filters and gave them enough flexibility.
There could be some way to share and reuse those filters maybe built on top of existing social networks as well so you won't miss out anything.
I would be able to switch to patio's feed for a week then someone else and check what they are reading.
Could work as a subscription model. Content creators or influencers can sell their feed filter as a service. What they see from twitter, reddit or hn, you can see it too.
You don't need to build a new network - one called the Fediverse (mostly made up of Mastodon and Pleroma and powered by the W3C standard ActivityPub) already exists and lots of cool people are on there.
It is however sorely lacking a way to give silent Bobs more attention - who posts more just gets more attention. Luckily it doesn't have a business model that would prevent one from building something cool on top of it.
I think it'd be pretty cool to have a curation algorithm that the user can control.
Lists work fairly well. I keep a set of three, by priority. "Top" is about 20 people I care to hear from, who post either rarely or very high quality. "Medium" is much less selective, and "Voluable" are overactive. I usually have the first two visible, frequently with "Home" (all followed profiles) as the 3rd stream.
I add, move, & remove frequently.
Another interesting dynamic on services permitting blocking is that, over time, the set of variously blocked accounts builds. This can go either way (submitter blocks follower, vice versa), but tends to result in interesting graph partitions over time.
Anybody remember the name of that open source browser-based RSS / social media feed aggregator from a while back? It had a really unique (colourful but simplistic) visual style and IIRC the domain had a novelty TLD. It had support for RSS feeds as well as plugins (?) for different social media sites. As far as I remember, one of the main talking points the author presented was the idea of being in control of the content you consume.
> Fraidycat (https://fraidyc.at/) is a delightful browser extension for following people on the interwebs. It's an easy-to-use yet hackable, FOSS, privacy-respecting, and experimental next-generation feed aggregator and reader. Not every platform offers RSS, so Fraidycat tries to build those bridges for you, scraping and packaging it up for you automagically. It aims to enable you to tailor your own feeds, priorities, and timelines from across the web in your browser.
> Instead of being beholden to cycling through endless applications and platforms to follow people: use Fraidycat to simplify, defragment, and take ownership of your window into the lives of others on the web. Fraidycat is afraid of what the web has become (and is becoming), and it's savagely fighting back.
> I want to bypass the treadmills and middlemen which aim to commodify my attention span and the pipelines between me and others. I don't want my browser to become just some surveilled thin-client dehumanizingly displaying a walled-garden web engineered by whatever an oligarchy of corporations chooses for me. Fraidycat allows me to more independently use my resources to track and represent people on the web for me. I'm convinced it gives the power of surfing and taking the pulse of the web back to the user.
> You own the aggregation process with Fraidycat because it runs in your browser. Political and technological autonomy requires turnkey tooling for users to actively shape the algorithms which pick out what they find salient on their own terms and devices. We have to shape our own filter-bubbles and how we feed ourselves information. Fraidycat is the kind of a lens-crafting tool we need.
I remember that for FB there is a "switch" that ends with "/chr" on the URL and that forces the posts to be displayed in untainted chronological order (newer on top). I haven't used FB for at least 5 years so that may have changed. I remember this well though because I had a bookmark on my FF toolbar with that exact URL so avoid the shitty manipulation of FB.
(I still believe that FB is cancer)(only came to read for Silent Bob!!!)
This same effect is in use through our email system. If you send heaps of email to each and every big provider (100 per day to even be registered by Microsoft's Outlook servers) then you can do something about no longer being sent to the spam folder automatically.
However if you're like Silent Bob and send out very few emails from your own self hosted server, you always get treated as though you're the next source of a major spam attack, despite having the same server and same email send rate for a decade.
As a dev I understand how this happens: it's all too easy to filter out the statistically insignificant. But we shouldn't.
NewsBlur has a nice feature called "infrequent site stories" that holds items like this. It's so much more useful to me than popularity-based measures.
I think the core of the issue is automated curation. There are several different ways to curate a collection. It isn't feasible to have someone curate things for you and you don't have time to do it yourself. Facebook/Twitter/etc has opted for majority curation: for the majority to define what is relevant to the rest of the users.
There needs to be a revolution in curation of content. Music suffers the same fate until we find a solution.
> In normal social interactions, when a person doesn’t talk much, more weight is often given to their words when they do.
I haven't found this to be true at all. Talking too little means your words don't get much weight in my experience.
This is not true for Silent Bob and the example band director because they have roles that command some amount of attention to begin with -- you're paying some attention to them when they're not talking.
In normal social interactions, in my experience, if you're very quiet then people just don't pay a lot of attention to you, and when you speak up it's not actually that jarring or novel. You haven't earned extra attention by not participating as a peer, you've earned less.
And if what you have to say is not something they already agree with they can easily go back to not spending their attention on you as soon as you stop talking.
It depends who the audience is. I definitely listen more attentively to less talkative people. I think the quality of social media would improve if they built that into their algorithms.
Also I think it may explain why people like me don't spend much time on social media. Too much BS. It's like watching too much TV; it dulls the mind.
When people spend most of their time expressing their ideas instead of forming and researching their ideas, their ideas are just not as good.
Is there possibly a selection bias at play here? Where you remember when a quiet person said something really meaningful, but the many times it didn't resonate, you immediately forgot about it? I'd be interested in studies on this material because I'm sure it's hard to quantify on a personal level.
> You haven't earned extra attention by not participating as a peer, you've earned less.
That's an interesting interpretation. I guess it depends on the social milieu. In mine, people are attentive to what Silent Bobs say. Very often it's an interesting piece of information because it tends to be different from the usual chatter.
My experience is that the too-talkative person is either discrediting more and more oneself ("leaving no doubt" about it) or it bores people away completely from the subject/platform/conversation at hand.
It's like, at some point, you leave your noisy/opinionated uncle speak in his corner at the family gathering. You don't mind him anymore.
Facebook went that route to me: what started to show up more and more from family and friends bored me away more and more, to the point I left. I prefer to pay the price to have no noisy-news and lose some "relatives" to only get the meaningful stuff when it matters. Left 2 years ago. No regret.
Twitter went the same route, about the time when the platform gained more public interest around 2015. Removed from my phone. Barely checking it now.
I tried Instagram alternatively. It's a pure marketing platform. Vaguely interesting as a pinterest-like journal.
It gets me back to blogs. Things that take more time (to setup/write, but also to find and read), bear more fruit.
I have discovered that Facebook can be manageable only if you are extremely diligent about unfollowing people whose posts you don't care about. You can unfollow while still remaining "friends". I have been really good about it and there are now at most 15-20 people whose posts actually end up in my news feed. Sometimes nothing new will show up in my news feed for several days, which is really nice. I think joining good, high quality groups of a not-too-big size is also a way to make Facebook a valuable experience. I was 100% off of facebook for a long time and even though I denied it for a long time, my social life suffered for it. I now see it as a necessary evil that can be not so bad with some effort.
Surely it isn't any great surprise that the algorithms on advertising funded sites are optimised to get the consumers to view and/or act upon the greatest number of adverts, rather than achieve any objectives which would run counter to this, such as finding a small amount of meaningful content. And from that premise, surely it isn't a great leap to imagine that the solution is to break away from the advertising funded model.
36 comments
[ 2.9 ms ] story [ 84.4 ms ] threadAnyone want to build it with me?
[1] https://wt.social/
At the core, I would do this by building a network graph of users who I've upvoted and users they've upvoted, and so on maybe a few levels deep. If anyone in that group of users has upvoted a comment, I want it on top - sorted by how many degrees of separation there are between me and the upvoter.
The very simplistic, single-faceted comment rankings that most places implement right now is, including Hacker News, is simply brain-dead IMO. I've brought this up before and 100% of the time, one of the responses I get is "sounds like a good way to build an echo chamber". It's not a bad thing actually. I think it more accurately models real life where we don't hang around with a bunch of people that we find disagreeable, even if they are in the majority and even if they're "right". With a system like the one I propose, everybody can feel comfortable, say what they want and things will most likely be way, way easier to moderate if they don't just moderate themselves.
er may be because of ... markets,democracy :). The majority have been trained to believe that the majority view is good.
If your users' needs are important to you, then your users need to be your customers - i.e. they need to be paying you.
If someone else is your customer, then their opinions and needs become more important than your users' opinions and needs.
The problem with Facebook is that advertisers are their customers and therefore they're the people that Facebook needs to keep happy. Providing a great experience for their users is secondary.
As users we get annoyed by this, but we're stuck with it until Facebook starts giving us the option to pay for our FB accounts. Which Zuck promised he'd never do.
So for your new shiny social network, the important question, as always, is: who's going to pay for it?
A protocol by itself indeed doesn't need a business model but clients for the protocol do need one.
They still don't need one, there can be many. Even if most people still end up with the gmail model, it's meaningfully different, because the switching cost is lower. You don't have to convince everybody you know to switch too.
Now suppose you're trying to solve the original problem. You've got feeds from a hundred users and you want a way to filter them yourself. If any service exists that provides this, using any business model, you can use it even if hardly anybody else does. You can also develop your own client that does this or commission someone to add the feature to an existing client.
More than that, if this is a feature many people want, now the ad-supported providers will need it too, because now they have to compete for users with the providers who are treating you as the customer instead of being insulated from competition by the network effect when they own the entire network.
And that's not going to be the norm. Most people don't want to self-host their email server, or even pay for it. They're happy trading their privacy for free gmail, and most don't even realise that this is a problem.
So as Nextgrid pointed out: that shifts the problem one step down the ladder. A commercial social client requires someone to pay for it.
Feeds in social media are severely limited for when you need to filter the content you want. What if you let people program their own filters and gave them enough flexibility.
There could be some way to share and reuse those filters maybe built on top of existing social networks as well so you won't miss out anything.
I would be able to switch to patio's feed for a week then someone else and check what they are reading.
Could work as a subscription model. Content creators or influencers can sell their feed filter as a service. What they see from twitter, reddit or hn, you can see it too.
I think it'd be pretty cool to have a curation algorithm that the user can control.
Currently I built a few simple aggregators:
https://pixelfed.club
https://mastodonia.club
But I am thinking of adding features to filter and surface, high signal/quality content.
I add, move, & remove frequently.
Another interesting dynamic on services permitting blocking is that, over time, the set of variously blocked accounts builds. This can go either way (submitter blocks follower, vice versa), but tends to result in interesting graph partitions over time.
Edit: Found it! https://fraidyc.at/
From the author:
> Fraidycat (https://fraidyc.at/) is a delightful browser extension for following people on the interwebs. It's an easy-to-use yet hackable, FOSS, privacy-respecting, and experimental next-generation feed aggregator and reader. Not every platform offers RSS, so Fraidycat tries to build those bridges for you, scraping and packaging it up for you automagically. It aims to enable you to tailor your own feeds, priorities, and timelines from across the web in your browser.
> Instead of being beholden to cycling through endless applications and platforms to follow people: use Fraidycat to simplify, defragment, and take ownership of your window into the lives of others on the web. Fraidycat is afraid of what the web has become (and is becoming), and it's savagely fighting back.
> I want to bypass the treadmills and middlemen which aim to commodify my attention span and the pipelines between me and others. I don't want my browser to become just some surveilled thin-client dehumanizingly displaying a walled-garden web engineered by whatever an oligarchy of corporations chooses for me. Fraidycat allows me to more independently use my resources to track and represent people on the web for me. I'm convinced it gives the power of surfing and taking the pulse of the web back to the user.
> You own the aggregation process with Fraidycat because it runs in your browser. Political and technological autonomy requires turnkey tooling for users to actively shape the algorithms which pick out what they find salient on their own terms and devices. We have to shape our own filter-bubbles and how we feed ourselves information. Fraidycat is the kind of a lens-crafting tool we need.
(I still believe that FB is cancer)(only came to read for Silent Bob!!!)
Now it would be awesome if FB lets me filter my news Excel Data Filter style.
However if you're like Silent Bob and send out very few emails from your own self hosted server, you always get treated as though you're the next source of a major spam attack, despite having the same server and same email send rate for a decade.
As a dev I understand how this happens: it's all too easy to filter out the statistically insignificant. But we shouldn't.
NewsBlur has a nice feature called "infrequent site stories" that holds items like this. It's so much more useful to me than popularity-based measures.
There needs to be a revolution in curation of content. Music suffers the same fate until we find a solution.
I haven't found this to be true at all. Talking too little means your words don't get much weight in my experience.
This is not true for Silent Bob and the example band director because they have roles that command some amount of attention to begin with -- you're paying some attention to them when they're not talking.
In normal social interactions, in my experience, if you're very quiet then people just don't pay a lot of attention to you, and when you speak up it's not actually that jarring or novel. You haven't earned extra attention by not participating as a peer, you've earned less.
And if what you have to say is not something they already agree with they can easily go back to not spending their attention on you as soon as you stop talking.
Also I think it may explain why people like me don't spend much time on social media. Too much BS. It's like watching too much TV; it dulls the mind.
When people spend most of their time expressing their ideas instead of forming and researching their ideas, their ideas are just not as good.
That's an interesting interpretation. I guess it depends on the social milieu. In mine, people are attentive to what Silent Bobs say. Very often it's an interesting piece of information because it tends to be different from the usual chatter.
It's like, at some point, you leave your noisy/opinionated uncle speak in his corner at the family gathering. You don't mind him anymore.
Facebook went that route to me: what started to show up more and more from family and friends bored me away more and more, to the point I left. I prefer to pay the price to have no noisy-news and lose some "relatives" to only get the meaningful stuff when it matters. Left 2 years ago. No regret.
Twitter went the same route, about the time when the platform gained more public interest around 2015. Removed from my phone. Barely checking it now.
I tried Instagram alternatively. It's a pure marketing platform. Vaguely interesting as a pinterest-like journal.
It gets me back to blogs. Things that take more time (to setup/write, but also to find and read), bear more fruit.