Conversation

Notices

1. MMN-o ✅⃠ (mmn)'s status on Tuesday, 05-Sep-2017 17:35:37 EDT MMN-o ✅⃠

   • Sean Tilley
   @deadsuperhero To be replaced with what, pseudoprivacy?
   • Christopher Lemmer Webber (cwebber)'s status on Tuesday, 05-Sep-2017 18:03:15 EDT Christopher Lemmer Webber

     • Bob Mottram
     • Sean Tilley

     @bob I think there are quite a few benefits that GNU Social could get by supporting AP. I hope there isn't too much "burn" by seeing it as a "Mastodon vs GNU Social" battle. I tried pretty hard to avoid that; @mmn and @deadsuperhero both at least can verify that I did try to get GNU Social and everyone else involved. I know that's not an obligation for GNU Social to adopt AP, not saying it is, just saying healing the "fractured federation" has always been a large reason for me working on AP.

     MMN-o ✅⃠ repeated this.
   • MMN-o ✅⃠ (mmn)'s status on Tuesday, 05-Sep-2017 18:37:58 EDT MMN-o ✅⃠

     • GNU Social
     • Christopher Lemmer Webber
     @cwebber The only reason !GNUsocial doesn't have #ActivityPub yet is because I have a fulltime job and noone else has been up for the task .]
     
     Though I'm pretty sure it'd still just be the 100% public parts of AP that would be used/promoted, as I'm pretty much convinced there's no such thing as privacy in the social sphere anyway and anyone using "private" communication in an environment like !GNUsocial or #Mastodon is fooled either by the platform, the administrator or other users. (anything accessible via a web browser isn't made for privacy)
   • MMN-o ✅⃠ (mmn)'s status on Tuesday, 05-Sep-2017 18:40:10 EDT MMN-o ✅⃠

     in reply to

     • Christopher Lemmer Webber
     @cwebber Indeed you and others involved with AP spec development have been very inclusive, inviting and encouraging. I very much appreciate your (and everyone else's) efforts that have been put into #ActivityPub.
   • MMN-o ✅⃠ (mmn)'s status on Wednesday, 06-Sep-2017 04:11:46 EDT MMN-o ✅⃠

     • you weary giants of 🏻 and ⛓
     • Christopher Lemmer Webber
     @nightpool @cwebber Whether e-mail is used or not is more a question of legacy than that it would respect anyone's privacy.
     
     My argumentation is primarily that an environment where the only difference between publishing your sensitive posts/pictures privately or publicly etc. is the value of some checkbox - then it's not appropriate for private communication. Even less so when that checkbox in practice is on a remote server governed by a remote admin.
     
     E-mail clients, for comparison, don't have a "share this post" button that potentially goes worldwide. Public mailing lists are explicitly opt-in and there's no real notion of a "public e-mail inbox".
     
     Federated social webs however don't work as they're dreamt about without eventually sharing information, metadata and posts to larger and larger - less controllable - groups of people.
     
     Centralised systems can control this more - but once you go federated then the end station you will reach is full-on public socialising.
   • MMN-o ✅⃠ (mmn)'s status on Wednesday, 06-Sep-2017 04:18:10 EDT MMN-o ✅⃠

     in reply to

     • you weary giants of 🏻 and ⛓
     • Christopher Lemmer Webber
     @nightpool @cwebber Essentially, I would like to have the discussion of "what happens when I click repeat and the post is private but I want to share it with the world?" to be somewhat settled among the _users_ (not us technicians) who will get baffled at the simple post privacy switcharoo.
   • MMN-o ✅⃠ (mmn)'s status on Wednesday, 06-Sep-2017 04:19:25 EDT MMN-o ✅⃠

     in reply to

     • Jabber and XMPP related group
     • you weary giants of 🏻 and ⛓
     • Christopher Lemmer Webber
     @nightpool @cwebber Oh, and for modern "e-mail but with privacy" I would say !xmpp's solved all of the issues at hand.
   • MMN-o ✅⃠ (mmn)'s status on Wednesday, 06-Sep-2017 04:53:20 EDT MMN-o ✅⃠

     • platano
     @platano Very true. My point is however that those things (leaks) can never be reduced to zero due to people being people. However, I don't want to see a system where these things can happen by accident. Compare with the policy changes at Facebook, where suddenly things you _thought_ you had set to private suddenly weren't because Facebook's idea of a "friend" (or what a friend's friend could see, etc.) has changed drastically a couple of times afaik.
   • MMN-o ✅⃠ (mmn)'s status on Wednesday, 06-Sep-2017 08:44:00 EDT MMN-o ✅⃠

     • Eugen
     @gargron While it's obvious I would code it like that, my argumentation is that any evil person anywhere could set up an evil node and users would have no idea...
   • MMN-o ✅⃠ (mmn)'s status on Thursday, 07-Sep-2017 05:57:25 EDT MMN-o ✅⃠

     • you weary giants of 🏻 and ⛓
     • Christopher Lemmer Webber
     @cwebber @nightpool web-likeness = public space. Web-unlikeness = privacy capable .]
   • MMN-o ✅⃠ (mmn)'s status on Thursday, 07-Sep-2017 06:01:28 EDT MMN-o ✅⃠

     • Christopher Lemmer Webber
     @cwebber Unless it was a feature with this certain host. Email is however considered like delivering via envelopes while social web platforms are like bulletin boards or at least relatively large group meetings.
   • MMN-o ✅⃠ (mmn)'s status on Thursday, 07-Sep-2017 17:47:58 EDT MMN-o ✅⃠

     • Jabber and XMPP related group
     • Christopher Lemmer Webber
     @cwebber Then again, the "expected usecase scenario" with !XMPP is like with e-mail. There's no UI (or expectation of any UI) that has a button which says "share this with the world instantly and publicly". Or any kind of accidental tagging-someone-into-a-conversation-and-thus-maybe-letting-them-get-some-conversation-history issues.
     
     I'm not talking technical and crypto privacy, I'm talking about how ordinary people are expected to use the communication methods. Which is an issue if 90% of the UI is designed for non-private data.
     
     I'm sorry if I'm repeating myself, I just want to be as accurate and explicit as possible in my commentary.