Do we need a new social media paradigm?

[recently_folded]

This thread is for thinking outside the existing boxes, or using the existing boxes in new ways. Let's start out with greywash's post the idea of a network of persistent fannish apothecary chests of holding.

Comments

[oulfis]

I was joking with greywash earlier that we're kind of hitting the problems that I'm examining in my dissertation -- that our problems kind of lie in linkrot. This is what I said to her in chat:

I've been playing around with having a personal website

partly to brush up on HTML/CSS and partly just to have a space totally by my own rules

so far it's BASICALLY nothing, just two poems, but it seems like a good place to put, like, my "uncomfortably sincere guillotine advice" and "personal reference for how to make a library look cool"

but even that makes me think about how helpful a, like, anti-linkrot system would be
if I put the guillotine advice on my own site, and mirror it to DW (as I plan to do), and then DW goes down and also I reorganize my site so the URL changes...

I would be totally willing to update someplace with some "btw this is all the same post" info to redirect links
rankly this is the problem that my dissertation is beating its head against in re: 18thC digital archives

oh!! do we need -- Linked Open Data Of Our Own??

I haven't really finished thinking that through, but here's a vaguely useful-looking primer on LOD: https://www.ontotext.com/knowledgehub/fundamentals/linked-data-linked-open-data/

A related phrase to google is The Semantic Web: https://www.ontotext.com/knowledgehub/fundamentals/what-is-the-semantic-web/

That's DEFINITELY not a good source for that info (it looks for-profit so I don't trust it) but this is The Good Stuff over in digital archival studies so it would be good to think about how fandom can get in on it.

[prettyarbitrary]

Okay, so having read the fannish apothecary thread, it sounds an awful lot like WorldCat, except for fandom crud instead of library crud. Essentially, we're talking a library catalog for the internet.

[satsuma]

i have no idea what this would look like technically, but it’s reallt interesting & def merits further discussion

I’d kinda started from the same place as you (redundancy is good) but then sort of went in the oppositr direction and ended up at “we should prioritise building in robust RSS support and personal backups into whatever we end up doing” like obviously fic is a little different than art but I LOVE that ao3 just lets me download a mobile file of literally whatever. And being able to use RSS to subscribe to non-Dreamwidth stuff through Dreamwidth is also great (or being able to give people who dont use dreamwidth a link to my blogs RSS feed and telling them to plug it into their own RSS reader)

[recently_folded]

I'm having a little difficulty articulating just what distinction I'm seeing here, but I do see a difference between a story meant to be read and simple conversation (since meta already has a place on AO3 as well). Do all fan artists want their work downloaded and kept by others as part of the enjoyment process? When they may be trying to sell it for needed income/fund-raising? It's not entirely apples/oranges, but definitely two varieties of apple.

So while I'm with you on RSS, which I still think is the handiest thing ever, and while I think that conserving fandom as a whole is the goal of these discussions, I'm not sure that relying upon this kind of mechanism provides the collective durability we really need. I know that runs up against the "everybody who cares participates in archiving" aspect of the P2P conversation happening in another thread, but that's my concern here. I think it's a fine feature to build in; I don't think it's a feature to rely upon more broadly, however. If nothing else, RSS is still something accessible only to a fairly geeky community and not the vast bulk of fandom who lives on devices only.

[prettyarbitrary]

In the past week, I've been catching up real fast on the federated/distributed internet. Gins brought it up in the 'fannish apothecary' post linked above.

The federated internet (aka the distributed internet, aka the social web) is the concept of a decentralized internet where, instead of a single platform--say, Facebook, or Reddit, or Asana--you have a piece of software that anybody can take and install on their own machine to create an instance of that platform. And then those instances can find each other and talk to each other, and users on them can talk across them.

Take email. You have a Gmail account. Someone else has a Yahoo account. But Yahoo and Gmail know how to talk to each other, because they both use the 'protocol' for email.

In the same way, in a federated social network, some enterprising folks snag copies of the code and install it on their own web servers. Other people, less enterprising but interested in participating in the community, join those servers, forming communities. We now have oh, say, 10 copies of Dreamwidth. You set up house on one of those copies, but it doesn't matter because you can talk across those instances--because part of the code is that they all know how to talk to each other and how to find each other. It's just a matter of having a 'home address' so to speak, so that the code (and people) know which installation to go to in order to find you.

For example, I don't know who has messed with Mastodon. Mastodon IS Twitter. Except anybody can build and own their very own Twitter (Twitter the site, not just an account). So on Mastodon, my username is @prettyarbitrary.arbis-mastodonserver--note that it has two parts, both my name and my home-base server (I set up my own installation, because I'm crazy like that). Now, you're my buddy and you have the username @zeldalover.flippers-mastodon (it's not YOUR server, because you don't know or care to know how to run a Mastodon server; you signed up on a friend's installation, and they've got about 30 other people who use it along with you). We can talk to each other because when I make a post, Mastodon says, "Ah, zeldalover is susscribed to read this. Wait, where's zeldalover? Oh, this says they're over on flippers-mastodon. I'll just toss it over there and flipper-smastodon will be able to hand it off to the right user."

And then also, because I own my own Mastodon, and flipper owns THEIR own Mastodon, a bunch of things happen:
1: Nobody has to foot the bill for a $200,000/year server farm, the way the AO3 does (I love them but they're terrifying)
2: Nobody has to organize god's own moderation squad, because we're all spread out into loads of small communities, and it's way easier to manage a 30 person community than, like, all of Tumblr
3: Nobody can come along and shut down the whole social network, because no single person or even small group OWNS it all. If somebody took out flippers-mastodon, I could throw open my doors and say, "Come on over here, y'all, I'm still up and running."

It ALSO means that the rules of behavior could vary enormously from one server to another. So if we're over on my server posting gross but legally neutral monster porn (look, I have my kinks, okay), but...I'm fond of imaginary flipper, now, I can't pin this on them...so let's say GrunterTheMRAGuy is over on HIS server, wheeling and dealing in child porn with his cesspool buddies (flipper and I both blocked his entire fucking server wholesale, because GTFO, scumbag). Grunter gets faceplanted by the FBI (and nothing of value was lost), but none of the rest of us are held accountable because we aren't even on the same machines as he is.

Depending on the network, we might even be able to build in the ability to make backup mirror copies of peoples' personal accounts. Maybe flipper and I mirror each other's stuff, so if flipper goes down, you guys can all come over and copies of all your info will still be there waiting for you. Nobody even lost a blog, yay!

But the real power of this setup rests in layering. If we have a bunch of these things going, what if we built a way for them to talk ACROSS platform types? So, let's say, we have a distributed system on which we can set up our core personal profile (and could this, perhaps, could be Gins' 'fannish apothecary chest?'). This is where a copy of all our information lives. We attach it to a set of OpenID login credentials. OpenID doesn't own our profile, and if we wanted to change names, we could detach that profile from that OpenID and move it to a new one. We use that OpenID to log in to our distributed social platforms: Mastodon, say, and also, distributed-Dreamwidth. Neither of those own our profile, OR our login credentials. BUT, because they send our info back to our profile, and also pull info from our profile, we can copy our posts from Mastodon over to distributed-Dreamwidth.

Many of these technologies already exist. OpenID is well-established (in fact it's on version 3, OpenID Connect). There are some early attempts at distributed profile-hosting out there. There are LOADS of distributed social platforms, including Twitter-clones (Mastodon), Reddit-clones (Steemit), Youtube-clones (PeerTube), and stuff that offers an unholy mix of different functionality. Wordpress--the set of code that you can install anywhere, not Wordpress.com, which is just one company's commercialized installation of it--is an early attempt at distributed blogging that's been around for ages.

As far as I know--and I may well be wrong, I'm still catching up on recent developments--we don't yet have a good way to talk ACROSS all these things. But W3C, the folks who design and organize the internet, DOES have a set of standards for building this stuff. It's called ActivityPub, and it was released at the beginning of 2018. This has the potential to kick development of this kind of technology forward at a faster pace, because now everybody can basically agree on what they're fundamentally doing.

[prettyarbitrary]

Also, if this stuff is already more or less taken for granted in this conversation and I'm approaching it starting several steps backward from everybody else, I apologize. I've been out of the loop on the web development side for a while, and I'm hurrying to catch up on the state of things.

[recently_folded]

No worries on that score, please. You're not the only one just now getting their head above water after backing up/populating new accounts/coming to grips with new community standards. And I think there's some merit in restating problems, concerns, and possible solutions various ways, just because we don't know precisely what will light the next spark of "aha! we can do [something]" that sets us in motion. So please, step forward, step backward, take lots of looks. I think that's what it will take to really explore these topics.

[altilis]

I read greywash's post about the "fannish apothecary" and they are right on the nose about the answer being some sort of wiki. I'm still catching up on the current situation and what capabilities that we're looking at, but I wanted to quickly come in and suggest: we should look into the already-existing, but nascent, idea of a Federated Wiki.

I'll elaborate on this more, soon, no later than this weekend hopefully.