Using ActivityPub as a common protocol for sharing social media across different open platforms seems like it could be a real killer feature for federated platforms.
With closed platforms like FaceBook, Twitter, etc. the owners of the platforms actively work to prevent users from sharing media with other platforms. Meanwhile, open federated alternatives are starting to embrace integration between them. This is a huge benefit from user perspective, as it means you can primarily use your platform of choice, but you can still interact with users of other platforms.
It’s really nice to see that a federated protocol has finally emerged, which projects are starting to use. The biggest weakness of federated systems, it seems, is when they inter-server protocol fragmentation happens, preventing actual meaningful federation to happen.
Now one must only hope (or help) that despite the decentralized infrastructure, these networks don’t centralize, and in the worst case split off (e.g. Google with XMPP).
This is really cool, especially in light of Youtube’s recent politically-motivated bans of certain kinds of videos depicting firearms. This is exactly the sort of technology that the web needs to fight corporate censorship of unpopular content.
In some sense, that is it’s downside too, isn’t it? Not only really bad content like misleading propaganda/conspiracy stuff, gore or mobbing, but also unregulated advertisements and clickbait would basically have no inherent opposition. All these things would contribute to a lower quality of the platform, unless there were a way to filter (ie “censor”) content. And assuming there that services like these have a tendency to centralize, the whole effect could be reversed.
You can moderate your instance just like any other service, you also have control with who you federate with. Moderation in my anecdotal experience in the federation tends to be better than their corporate counterparts, not worse. Sure there are some really repugnant instances, but they’re pretty clear about their moderation policies and my instance doesn’t federate with them. So it doesn’t appear to be a dualism, rather you have more flexibility.
Actually from that specific angle it’s not so cool, there’s a risk of PeerTube instances becoming a sort of Voat where the worst sort of extreme content pop up.
Oh man, this almost exactly scratches the itch we (friends and I) have for a project. Unfortunately, the use of the AGPL license makes me somewhat unwilling to use it. :(
Yes I think excluding people who want to profit without sharing their changes was definitely a prerequisite to funding by the non-profit who funded this. There’s nothing stopping you though from creating your own that is whatever license you like. Webtorrent is MIT. Personally I’m a fan of the first one coming in to be AGPL, it prevents the kind of embrace, extend, extinguish pattern. Then others can make MIT, BSD, or whatever and if they get extinguished we still have something to fall back on.
I was just asking @friendlysocks. Personally I am fan of strong dopyleft and the AGPL but always interested in other peoples reasoning and objections against them.
As written, a sufficiently grumpy reading suggests that you need to provide sourcecode for everything that touches a packet that flows through the network. If that isn’t the case, there is no real benefit to using this over basic GPLv3 or similar copyleft.
A compilation of a covered work with other separate and independent works, which are not by their nature extensions of the covered work, and which are not combined with it such as to form a larger program, in or on a volume of a storage or distribution medium, is called an “aggregate” if the compilation and its resulting copyright are not used to limit the access or legal rights of the compilation’s users beyond what the individual works permit. Inclusion of a covered work in an aggregate does not cause this License to apply to the other parts of the aggregate.
Like, it almost looks like “oh, so including this in my stack is okay”, but you read it again and it’s like “wait, shit, am an ‘extension’ of a covered work? what constitutes a ‘larger program’? what about containers?” and the whole thing just becomes ornery.
Using ActivityPub as a common protocol for sharing social media across different open platforms seems like it could be a real killer feature for federated platforms.
With closed platforms like FaceBook, Twitter, etc. the owners of the platforms actively work to prevent users from sharing media with other platforms. Meanwhile, open federated alternatives are starting to embrace integration between them. This is a huge benefit from user perspective, as it means you can primarily use your platform of choice, but you can still interact with users of other platforms.
It’s really nice to see that a federated protocol has finally emerged, which projects are starting to use. The biggest weakness of federated systems, it seems, is when they inter-server protocol fragmentation happens, preventing actual meaningful federation to happen.
Now one must only hope (or help) that despite the decentralized infrastructure, these networks don’t centralize, and in the worst case split off (e.g. Google with XMPP).
There’s talk of allowing GitLab events to be published with ActivityPub as well: https://gitlab.com/gitlab-org/gitlab-ce/issues/44486
This is really cool, especially in light of Youtube’s recent politically-motivated bans of certain kinds of videos depicting firearms. This is exactly the sort of technology that the web needs to fight corporate censorship of unpopular content.
In some sense, that is it’s downside too, isn’t it? Not only really bad content like misleading propaganda/conspiracy stuff, gore or mobbing, but also unregulated advertisements and clickbait would basically have no inherent opposition. All these things would contribute to a lower quality of the platform, unless there were a way to filter (ie “censor”) content. And assuming there that services like these have a tendency to centralize, the whole effect could be reversed.Edit: I misunderstood the project.
You can moderate your instance just like any other service, you also have control with who you federate with. Moderation in my anecdotal experience in the federation tends to be better than their corporate counterparts, not worse. Sure there are some really repugnant instances, but they’re pretty clear about their moderation policies and my instance doesn’t federate with them. So it doesn’t appear to be a dualism, rather you have more flexibility.
Actually from that specific angle it’s not so cool, there’s a risk of PeerTube instances becoming a sort of Voat where the worst sort of extreme content pop up.
Oh man, this almost exactly scratches the itch we (friends and I) have for a project. Unfortunately, the use of the AGPL license makes me somewhat unwilling to use it. :(
What’s your trouble with AGPL? The obligation to share your changes, the patent clause or something else?
Yes I think excluding people who want to profit without sharing their changes was definitely a prerequisite to funding by the non-profit who funded this. There’s nothing stopping you though from creating your own that is whatever license you like. Webtorrent is MIT. Personally I’m a fan of the first one coming in to be AGPL, it prevents the kind of embrace, extend, extinguish pattern. Then others can make MIT, BSD, or whatever and if they get extinguished we still have something to fall back on.
I was just asking @friendlysocks. Personally I am fan of strong dopyleft and the AGPL but always interested in other peoples reasoning and objections against them.
Sorry, I was merely explaining why the license was chosen and couldn’t help myself in peanut gallery support.
As written, a sufficiently grumpy reading suggests that you need to provide sourcecode for everything that touches a packet that flows through the network. If that isn’t the case, there is no real benefit to using this over basic GPLv3 or similar copyleft.
If you look at the thing in question, there are tricky bits like this:
Like, it almost looks like “oh, so including this in my stack is okay”, but you read it again and it’s like “wait, shit, am an ‘extension’ of a covered work? what constitutes a ‘larger program’? what about containers?” and the whole thing just becomes ornery.
“Unfortunately, the use of the AGPL license makes me somewhat unwilling to use it. :(”
I’m collecting the objections to that one since I don’t see it mentioned often. What’s yours?