@bob For the #OMN being based on the #4opens am looking at Odata as the rest API core https://www.progress.com/blogs/odata-faqs-why-should-rest-api-developers-use-odata as this is an outgrouth of #atompub and #RSS and comes from a long process of open development #indymedia What do you think?
@Hamishcampbell @bob It looks like this is the relevant specification document:
http://docs.oasis-open.org/odata/odata-json-format/v4.0/os/odata-json-format-v4.0-os.html
I'm not aware of anything currently using the OData format and in the context of federated systems I think it would be better to stick with the ActivityStreams vocabulary, which is what ActivityPub is based on.
@bob @Hamishcampbell this is based on atompub which is the same thing activertypub is built on. Am format agnostic as long as its #4opens am not soure that activertypub implements the update data option, its not in any of the apps. For the backend am looking to build a huge distributed "lossy" database - the data soup were the apps are just laddles tto this bubbleing coldren. No resion you can use activertypub to dip in for norshment :)
@hamishcampbell @bob @Hamishcampbell It sounds like what you want is a distributed database with redundant storage. Tahoe-LAFS is similar to that in that some servers can go down but the data can still be reconstructed, but those things are usually immutable data only.
If you're using text files as a database then the old skool method would just be to rsync a bunch of servers. Syncthing would be a more modern way of achieving the same result.
@bob @hamishcampbell bit more dinamic than that - its a human network/database built out of flows of trust http://hamishcampbell.com/index.php/2020/03/05/what-is-the-omn/
@hamishcampbell @Hamishcampbell @bob The open web is both social and technical. Over its history we've learned a few lessons the hard way. If you create ungovernable social spaces then they will be colonized by the worst people on the internet.
Cryptography doesn't create trust (as the Bitcoiners often like to imply) but it does provide a framework within which it's possible for trust relations to happen.
@bob "open" to create trust and value. if people brake the trust the tools are all about moveing them into a commerty were there views can be trusted. If they have no vlaue then they will be organicly dropped from the network - spam is an example of this. The network is inherently lossy and leeky to cross the filterbubbles and keep the data from ossifying as our current silos do. Open has power that closed duse not. Yes open has dangurs we know well that closed duse not in anyway solve.
the exciting bit are the "flows of trust" bulding commertys. the tech and code are just tools.
* nothing new
* all #4opens
* good UI
@Hamishcampbell @bob this is of coures what mastodon trys to build on the work of twitter. BUT mastodon is based on individuals - am looking to build the same based on "flows" which can be anything. This should speak activertypub, RSS, activertystreams and odata etc. its just data in and out - what matter is who conects the flows - the social side of tech, and what these flows nurish as an outcome #indymedia
@hamishcampbell @Hamishcampbell @bob That sounds similar to the Hubzilla/Zap "channels". A channel can be an individual, but it can also be a group or a fanpage.
@bob what we have to build is the "trust network" and datasoup infrustructer. all the current apps becouse laddles and soup bowls.
What we are building is like the WWW its a #KISS user built sematic web runing over the current internets with existing distributed data and tools to use it. the key word is KISS which is a good start to a flertation ;)
@Hamishcampbell @bob
ie. its anacky in code - much like the #openweb was. its a social tecnlogicl project. not a tecnilogical social project. ie. the social comes first to clearly moive away from the #'geekproblem. but how to get the #geekproblem to implement something that is desined to move power away from them... thats the problem.