Conversation
Notices
-
@strypey @arunisaac @bob @mike I only see one side of the discussion, but OStatus should cover both of these things.
- clacke repeated this.
-
I'd day that Hubzilla is a significantly better system than gnusocial, enabling some real privacy especially if you host your own node. The only down side is that there are fewer people using it.
-
Have you read https://indieweb.org/monoculture ? According to that article, systems like #hubzilla, #gnusocial, #diaspora might classify as monocultures that fail to interoperate with other systems, and force everybody to run the same software on their servers. Thoughts on this?
PS: I am aware that gnusocial does have a level of interoperability with other similar projects. But, I have no idea about diaspora and hubzilla.
-
@arunisaac hubzilla came from Friendica, and Friendica interoperates with a lot of things. What was found was that the price for interoperability was a broken privacy model.
What's missing for all of these things is an open standard protocol for private and public communications. I think that was suggested at the very beginning of the Diaspora project (define a protocol first, then build something around it), but that's not the way they went.
-
The #indieweb uses #microformats such as h-entry and h-card to facilitate inter-operation between websites. I like the idea, and am beginning to implement it on my blog.
https://indieweb.org/POSSE
I notice that #gnusocial is marked up with these microformats. #hubzilla, I haven't tried yet.
-
@bob @arunisaac broad consensus on open standards for social apps is a wicked problem, because it has both technical and social aspects
-
@bob @arunisaac is a stewardship body for social web standards needed? See my blog comment when Diaspora got funded:
http://qttr.at/1ho4
-
@strypey @arunisaac I think there has been some attempt at a W3C standard, but I know that Mike McGirvin (Hubzilla) has no faith in that. Anything implemented needs to be able to define which users can see or interact with what content, otherwise you end up with systems which are public only but of no value for private communications.
-
@bob @arunisaac @mike I'm not convinced it makes sense to do both publishing and private comms in the same tool. Opposite requirements
-
@mike @bob @arunisaac I think we need web standards for social *media* (publishing) and separate standard(s) for social *networking* (comms)
-
@bob @mike @arunisaac the problem I see is that the publishing functions are, by nature, security weaknesses in the private functions
-
@mike @bob @arunisaac and the careful technical work required to protect privacy makes for kludgy publishing tools
-
@bob @mike @arunisaac to be specific; private coms tools with no publishing functions *cannot* make things public by user accident OR attack
-
@bob @arunisaac "Broken" is Mike's loaded term. His ideas of nomadic identity and encrypted delivery didn't work well with other systems.
-
@bob @strypey @arunisaac There is a WG, and they just released CRs of #MicroPub and #ActivityStreams2 and #ActivityPub is in the works.
-
@strypey @bob @arunisaac #micropub defines client-server interactions for the indieweb: https://aaronparecki.com/2016/08/23/2/micropub-cr
-
@bob @arunisaac @clacke I don't know that "better" is a good description. "Different" as in different focus (privacy, general site for a group vs social for individuals)
-
@strypey @bob @arunisaac #activitystreams2 and #activityvocabulary define social objects like people and actions, and how they fit together.
-
@strypey @bob @arunisaac #activitystreams2 https://e14n.com/evan/note/kDqyDcwySL-rBoVhYztj2A
-
@strypey @bob @arunisaac #activitypub defines client-server and server-server interactions for #activitystreams2 and covers access privacy.
-
@strypey @bob @arunisaac #activitypub https://www.w3.org/TR/activitypub/ might be a CR in two weeks: http://qttr.at/1hp1 (identi.ca)
-
@lambadalambda @strypey @mike @bob @arunisaac OStatus has no sense of a private conversation or message, only public messaging.