Ariadne Conill 🐰 (@ariadne@treehouse.systems)
social.treehouse.systems
i hate to say it, especially as the person who started IRCv3 in the first place, but there is literally no world in which i would deploy a new project on IRC. i can manage all moderation tasks on discord with terraform. that is something impossible to realize with IRC. any project to take back mindshare from Discord has to frame their strategy from this perspective. with Discord, or any other SaaS, you are dealing with a loss of software freedom, and that should be highlighted, but the solution is to provide a libre alternative that is competitive. IRC (and frankly Matrix) isn't that.

cross-posted from: https://lemmy.ml/post/957811

Ariadne Conill 🐰 @ariadne EN

i hate to say it, especially as the person who started IRCv3 in the first place, but there is literally no world in which i would deploy a new project on IRC.

i can manage all moderation tasks on discord with terraform. that is something impossible to realize with IRC.

any project to take back mindshare from Discord has to frame their strategy from this perspective.

with Discord, or any other SaaS, you are dealing with a loss of software freedom, and that should be highlighted, but the solution is to provide a libre alternative that is competitive. IRC (and frankly Matrix) isn’t that.

https://social.treehouse.systems/@ariadne/110199185608729068

Ariadne Conill 🐰 @ariadne EN

this has always been the problem i was trying to solve. with atheme, with ircv3, with all of it. how do we provide community plumbing that is usable and scalable?

but IRC failed to evolve fast enough despite all of those efforts, because people didn’t understand the real evolutionary threats.

the reality distortion field is a real threat to any free software project: it is very easy to become complacent, because the product is 70% of what is actually needed.

but as the world evolves, that 70% turns into 60% and then 50% and so on, while people resist the concept that product fit and focus are slipping.

everything is Fine™️ because everything is Free™️, and instead of focusing on the real threat (I have been saying that IRC would be eaten by proprietary services since the 2000s), people, thinking that everything is fine, actually, tend to focus on their little kingdoms rather than the big picture.

and so we slipped, and slipped, until eventually, IRC does 20% of what we want, and IRCv3 brings that to maybe 25%, and then when a rich charlatan buys the largest IRC network and ruins it, at least half the people still there who haven’t left yet realize that, upon having their reality distortion field shattered, actually 25% of what is needed perhaps isn’t the right thing, and they too move their projects to Discord or Slack.

this is a problem, and it needs to be fought, but any such fight needs to write IRC off as a loss and start over. this isn’t about “how do we win over the 1990s chatroom user”, it’s about “how do we win over the 2023 discord user.”

https://social.treehouse.systems/@ariadne/110199267432776043 Ariadne Conill 🐰 @ariadne EN

like, seriously, you have no idea how frustrating it was to try to mold IRC into a competitive product. i tried for a decade. i even worked on this as a full-time SRE for a while (Ustream really needed UnrealIRCd to be rewritten).

we even had some wins, for a while: the decline of IRC’s userbase was reversed and it even grew for a while.

but for the most part i had the pleasure of advocating that IRC developers do not do stupid shit, like add spying features (looking at you InspIRCd m_invisible.so).

at the ecosystem level, the strong desire of IRC developers to do stupid shit for short-term gains in users, outpaced the desire to promote the health of the ecosystem, and add new competitive features that end-users would care about.

this is because projects cared far more about admin mindshare than user mindshare, and basically shows how the whole IRC mentality is doomed to failure.

community infrastructure projects have to be community focused, not admin focused. Rob Levin (the founder of freenode) used to derisively refer to the people who didn’t get this point as “traditional IRC users.”

https://social.treehouse.systems/@ariadne/110199317380320492 Ariadne Conill 🐰 @ariadne@treehouse.systems

incidentally, the fediverse is in a similar position, where it is threatened by the free software reality distortion field.

Mastodon isn’t good enough for the long run. we laugh at BlueSky, but it is a legitimate threat, and it could very easily wind up eating the fediverse. all they have to do is make it more palatable to the mainstream.

https://social.treehouse.systems/@ariadne/110199369743497066 Ariadne Conill 🐰 @ariadne EN

one last thing. the people who are suggesting FOSS chat alternatives to me.

you miss my point. you’re offering me oranges when Discord has offered me an apple.

Discord is not “just a chat platform.” I would describe it as an “integrated community management platform.”

It combines chat with other forms of community media: forums, for example, and a rich suite of AV capabilities.

this is also a symptom of the free software reality distortion field: the alternatives suggested may be sufficient for some usecases, but that doesn’t mean they cover the same niche as the product they are proposed as a replacement to.

it’s the concision of experience that has allowed Discord to have such great success in their efforts to eat IRC’s userbase.

  • _ed
    7·
    2 years ago

    Mastodon isn’t good enough for the long run. we laugh at BlueSky, but it is a legitimate threat, and it could very easily wind up eating the fediverse. all they have to do is make it more palatable to the mainstream

    I can see existing Fedi services adopting the blue sky protocol to stay relevant, once slicker platforms start running on it.

    • fmc01@mstdn.social
      4·
      2 years ago

      @_ed @indieterminacy Why would a slicker platform want to empower its customers to go elsewhere - I expect Twitter to starve it to death and the rest of the platforms cheering on. Mastodons strength is a committed group of people who manage the servers - they will need our respect and financial support. Blue sky is just that …. Blue sky.

      • ttmrichter@lemmy.ml
        4·
        2 years ago

        And here we see the seeds of IRC’sthe Fediverse’s irrelevancy being planted and lovingly tended.

        What you just said could have easily been put into the mouths of any number of IRC advocates 20 years ago. Where is IRC now? Remind me.

    • maegul (he/they)@lemmy.ml
      31·
      2 years ago

      Yea my negative “ahuh “ moment with the fediverse was that there’s a good chance it (ie ActPub and prob the current main platforms) is ephemeral and won’t really be successful past the mid term.

      Not insightful (most tech comes and goes). But once I saw enough evidence of flaws and distortion fields and somewhat typical OSS “diversity” and “culture policing” it was pretty apparent to me.

      Hope I’m wrong, because it’d be wonderful if healthy protocols replaced big social.

    • indieterminacy@lemmy.mlOP
      1·
      2 years ago

      I must admit, Blue Sky has been in my periphery.

      I glanced at the AT protocol but felt it wouldnt match my needs and values. I would expect that there is enough going on to provide enough momentum for that, so at least they can pivot into something else having learnt lessons.

    • indieterminacy@lemmy.mlOP
      1·
      2 years ago

      FWIW, I just read this comment by the user, snarfed - in the matrix room, Federated Devs: https://matrix.to/#/#fediverse-devs:matrix.org

      in practice I’m not sure how much it adds, since AP server admins still generally store and control users’ private keys, can sign with them, rotate them, etc this is one thing I like about Bluesky. it’s federated, but it has two levels of keys per user. servers have one level of private key, used for normal interactions. users have another layer that’s only used to switch servers, “undo” a malicious server’s recent activities with its user key, etc.

      https://atproto.com/guides/overview#account-portability

      The topic came up with-regards-to the complexity and poor interoperability of http signatures between fediverse instances.