The people behind it are transitioning from a services business to an open platform cooperative, with exciting vision and plans (unfortunately they had to do this transition quicker than planned).
Maybe you are interested to reach out to Mathew Mytka, as they are looking for the kind of expertise you bring to OERu, among others..
See for more background: https://greaterthanlearning.online/blog/shifting-the-short-term-focus/
@humanetech I've joined their platform... although I'm not seeing any evidence of their open source-ness... they talk a lot about openness and transparency, but not seeing any code....
Hi @lightweight sorry to be late in responding. Yes, you are right, but it is in the transition that these aspects will be addressed. Mat calls it 'Metamorphosis'. And what that entails will be up to those involved in the elaboration, and is effort open to anyone willing to join.
Mat will respond later, with his new fedi acount at @m3me when it becomes operational ;)
Spec conformance is not enough, besides format there's also the #linkeddata semantics that should match.
Various forum topics ongoing. Today I typed this on the subject: https://socialhub.activitypub.rocks/t/what-does-activitystreams-conformance-mean/1733/4
Do you have more info on avalanche attacks? Unfamiliar with the concept.
Yea, schema.org is mostly the whole SEO shebang.
I think for a long time to come fedi will use LD as-if it were plain json, and just add a @context property for good measure.
Most extensions will be closed vocabs anyway, hopefully well-defined, just standardized enough for decent amount of interop. No semantic webs, or anything near that any time soon :)
@humanetech @bob @m3me @lightweight By avalanche attack I mean that if a schema needs to be resolved then that may require going through a series of servers, similar to DNS. Changes to the top level schema on a single server can then have cascading effects throughout the whole network. So even though your system may be federated the interpretation and security (json signatures) of data may come to depend upon one or a small number of servers. It can be mitigated by cloning/caching, but there is then also the possibility of poisoning and which schema is the real one.
The idea of semantic markup isn't bad, and could be very useful. It's just that via schema.org Google gets to impose its particular ontology on the world. Some people around here may have experience with having inappropriate ontologies superimposed onto them.
A private instance for the Finkhäuser family.