Even though I failed, I benefited quite a bit by taking on something so ambitious, and whether Diaspora succeeds or fails, the people working on it will most like be able to say the same.
Didn't it occur to you that even if diaspora goes nowhere, the work these guys have been doing may be influential or even the foundation for something that will actually go somewhere in the future?
While "only" 4 years, Muhammad Ali sacrificed what would have been the prime of his boxing career as a conscientious objector. He was famous as a great athlete before, but transcended it after.
Alexander Calder was an engineer for about ten years before becoming one of the great artists of the 20th century.
Many other examples.
Any way, the goal is not to kill facebook. It is to experiment how a decentralized social network could look like, and provide a solid alternative to those who don't want to see their data analyzed. It's nothing more than that at the moment.
For the first point, our protocol is improving (see https://blog.diasporafoundation.org/43-our-federation-protoc...) and is the only one really production ready imo, and for the second point, the user experience is getting better at each release, with new features.
Frendica is another, I think, related protocol.
The problem generally is that the communities are small, diversified, hosting is a hurdle for virtually anyone, and individual instances can be finicky.
(On Mastodon -- a different technology entirely, but similar in concept -- I have two accounts on different instances, and since April have found that one or the other has been down, unavailable, or technically unusable for up to weeks at a time.)
Should some nucleating group decide that they were going all-in on Diasapora (or a compatible tech), that might make a difference. Meantime, everything seems stuck in slow-start mode.
The slow-start mode seems to be default mode for all new social media platforms while FB is dominating the space. I could see a service business rolling Diaspora out on-site to organizations. I think poaching users is fair game and should be perfected across all the platforms. Basically, sharing on FB links that are hosted on a Diaspora instance. So the act of sharing on FB increases exposure for the instance.
https://en.m.wikipedia.org/wiki/Diaspora_(social_network) citing The Federation: https://the-federation.info
That's ... on the order of Usenet ~1990 or so, per personal conversations with Gene Spafford a few years back. Where "OTO" could be 500k - 5m users. Much of Usenet was far smaller.
Microsoft did some studies on Usenet nodes and behavioural patterns in the early 2000s, and got some usage numbers out of that, though I'd have to dig to find them again.
(Funniest part of the requirements spec: it had to "appear to be" hosted on a Windows server...)
Also, its community does very much consist out of the more technical crowd, with a focus on privacy and it also helps if you don't mind the occasional free software activist shouting about. That's just the group that's most likely to sign up to it, which can be a blessing as even with so few people, you always find someone to talk about tech stuff, but it can at times also get somewhat old...
what is needed is a facebook e.V. with a strict data safety policy.
https://en.m.wikipedia.org/wiki/Registered_association_(Germ...
it would cost transparently what is needed to sustain the service.
Historically, neither of these tend to be sustainable.
Also government is nothing bad. A part of a good data safety policy is absolutely also including cooperation with law enforcement. If you think that your government is abusing its power then choosing super-crypto is not the right battle field - you have to become politically active.
data is controlled by:
a) yourself b) your hacker-friend c) a profit-oriented company d) a non-profit-oriented organisation
a and b are the restrictions diaspora faces.
Facebook is c - not good for well-known reasons.
d is an organisation responsibly handling your data.
no contradiction.
the organization though could in fact be based on Diaspora and maintain pods.
Also, ISTR from last time I played with it, there was no ability to create photo albums/collections. Does that exist now?
Without those two things I couldn't even begin to consider getting people to start using it.
No, not yet. See https://github.com/diaspora/diaspora/issues/1359
1. They market it like crazy in the hopes of getting something like a few million users
2. They publicly state that they will mine all metadata for the purpose of generating an "inference report".
3. Every other week they release an "inference report" that reveals a new, dangerous way the seemingly innocuous metadata can be used. Some examples would include a) accurately gleaning more private data from the metadata, and-- if enough people join-- b) using that data to subtly influence the behavior of the participants.
Outside researchers are given access to the process in order to audit it. Anything revealed in the inference report would be assumed to already be happening on larger commercial networks.
Users would remain as long as they believe the value of the inference reports outweigh the risk to them of using the network.
Edit: formatting
That's the problem. Nobody uses Disapora*.
I got an account a few years ago, but abandoned it after discovering there was nobody on there worth talking to.
Everybody want to see the results, rarely who will want to actually be there.
How do you incentivise it?
But ... isn't Diaspora trying to solve a too big problem? Ideally, shouldn't the decentralization-space be filled by smaller services? For example, in my perfect world, instant messaging and newsfeeds would be two separate projects. That way, not only is the information decentralized, but also the development of the decentralized web becomes more decentralized (in a way).
I dont want to be on a platform that can hold my friend hostage, is why I'm asking.
EDIT: If hub federation is needed for asynch messages, and unblockable (by hub owner) p2p used for chat when both users are online, then I guess that would be acceptable.
Would be a good kickstarter for generating sufficient network effect value to use in the first place, evolving the platform software through real world tire kicking, and getting it out to the wider world. And being github, as well as a sufficiently technical userbase, there's already a business model around freemium hosting which could be applied.
