I've come across a lot of people in the fediverse who appear to be using Mastodon, but I think would find the privacy model of Diaspora much more to their liking. If Diaspora ever has a change of heart and decides to federate over ActivityPub, and produce some user-friendly web videos etc explaining how the Diaspora UI works, I suspect they would have a huge uptick in users. I can certainly think of people I'd encourage to migrate there.

#Diaposra #ActivityPub #Mastodon

@strypey Very much agree. FLOSS protocols *need* to have interoperability as first-class citizen built in, unfortunately this just rarely happens at the moment, building more "walled gardens" rather than eliminating them. Looking at here too. 😟

@z428 @strypey I don't really use Movim, but I don't think throwing this shade at it is fair.

For almost a decade, Movim was using the only federated social network standard there was, and the only federated social network specification with more than one production implementation, until the W3C recommended AP in 2018. It was everyone else that wasn't catching on.

I think it can be forgiven for not yet having adopted AP, however cool I think it would be.

@clacke I'm not really offending here. I'm, like, more a bit concerned to see that repeated "reinvent-the-wheel" culture in FLOSS nowadays(?). We're blaming Facebook and Twitter to be "walled gardens" for business and control reasons but fail for ourselves to build solutions that aren't "walled" gardens yet for entirely different reasons... 😟


Β· Β· 1 Β· 0 Β· 1
@z428 @strypey I think the fragmentation is sad, but the huge difference is that the documentation and access necessary to bridge e.g. Movim and Mastodon is out there and anyone is welcome to do it, it's just that nobody has.

They're separate gardens and I'd like to see them grow together, but at least there's no wall of obfuscation, anti-features and terms of service to separate them, I can even follow Movim from Friendica, as it exports feeds.

@clacke Yeah, I see that too - but on the other side, this integration never works "seamless". This bridging and transformation always comes at the price of incompatibilities and limitations - and, like you said, it always initially boils down to "someone needs to do it". Looking at this seems the worst at the moment - in quite some cases (Diaspora <-> AP, Movim <-> AP, ...) nobody does it because people expect each other to do it, leaving it "practically walled gardens" because ...


I agree with @clacke that the federated gardens are not walled, but I agree with @z428 that there's a lack of well-engineered paths between them. In the case of Movim <> Diaspora <> AP, everyone seems to use the cop-out of accusing the others of NIH syndrome, as an excuse for not making any path-building efforts on their end.

@clacke I suspect part of the problem is self-taught developers who have learned a lot of their craft working on that one set of protocols, in that one codebase, in that one language. They have programming skills that dwarf my kindergarten level coding, but they don't yet have the software engineering mastery of the people who built the plumbing of the net in the 1980s/ 90s. So they use scorn towards other projects as a way to avoid admitting their lack of confidence with other toolsets.

@clacke if I'm right about that, then pointing out the UX problem to those project devs and negotiating with them is unlikely to help. Maybe what we need to do is recruit greyhairs of the generation that standardized on TCP/IP and the email protocols, to act as mentors to anyone willing to work on smooth two-lane bridges between any two federated gardens. Whether those mentees are part of either of the projects being bridged, or not.

@strypey Agree without (hopefully) sounding too angry here. This feeds my understanding, however, that in order to substantially change things, we might need to find an idea to drive forth "open"/"libre" alternatives yet in a *professional* way. Does Software Libre, do alternative social networks *have* to be built by hobbyists in their spare time for learning and fun? Or would it at some point require people who do this full-time with a profound set of skills and experience..? πŸ˜‰


> find an idea to drive forth "open"/"libre" alternatives yet in a *professional* way

You mean like ... Matrix ;)

@clacke 9

software meta, Matrix 

@strypey @z428 @clacke


Not at all like Matrix.

Matrix is ops hell.
Matrix is useability/ux hell.

If Matrix is a/the leading example of "professional" in open source, as an engineer I have have a few scathingly worded critiques of the definition, and will never be using it again.

I mean, have you *seen* the hoops you're jumping through just to install it?

Matrix as it exists right now is a cantankerous, half-assed proof of concept prototype with a few cute features, and it'll be interesting if they'd actually cut out the polycentric serverside bs.

It has potential, if they feature freeze, polish, and refine it aggressively for the next ten months.

But "professional", as-is?

Heck no.

software meta, Matrix 

@eryn Muahahahahah, well, at least I have to agree with that "ops hell" part. The best thing, however, to consider "professional" surrounding Matrix is the fact that they apparently managed to come up with a project management organization, somewhat well-crafted(?) goals and determination to reach that. Even that given seems more than other projects unfortunately try to achieve. 😐

@strypey @clacke

@z428 @strypey all I'm saying is movim and SΓ T got first dibs on that attitude πŸ˜‰

@clacke yes, but the AP networks have an equally valid argument, that their combined user base dwarfs the combined user base of Movim and SaT. So why should the AP devs bend over backwards to support an arguably obsolete standard.

But both arguments are a cop out. Trotting them out does nothing to improve the fragmented UX of federated networks, which contributes to people staying on the datafarms, which hurts all federated projects equally.


@strypey I agree. My most pressing pain with this is, maybe, on a meta-level: I don't care at all *who* exactly is responsible to build a bridge. I, however, complain in a rather bitter way that, at this point, apparently weird mechanisms work and keep people from doing the obvious, from doing what an early FLOSS idea should be expected: Stick together and get things together rather than everyone sitting on her/his own island and blaming neighbors for not wanting to bridge the gap. 😐


We're relatively new to the fediverse, if we were to setup a Tor or #i2p instance, would clearnet instances have the infrastructure to communicate with such a #decentralised internet?

Are there mechanisms in place to help protect the #anonymity of #Tor users (eg. traffic/random junk traffic only sent and received between instances every 12 hours at a universal time etc?)
@strypey @clacke

@dsfgs Boosting this as I am not really sure what's best to tackle that. πŸ™‚

@strypey @clacke

@dsfgs good questions, well above my paygrade. Maybe start a thread about this on socialhub.ActivityPub.rocks or
forum.privacytools.io ?

@z428 @clacke

@clacke ... those who would use such bridges might not have the skills to actually implement (let alone run) them by themselves. That's why quite often I'd feel better having compatibility a "first class citizen" in these protocols rather than completely following that dreaded "move-fast-and-break-things" approach. πŸ™‚


Sign in to participate in the conversation

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!