-
Notifications
You must be signed in to change notification settings - Fork 2
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
We need a vocabulary of Service Hub types #9
Comments
Or better, by some way (magic) the orchestrator can find out what the service type if of a Service Hub. This is also the preference of Herbert to keep the Event messages as light as possible. We could e.g. propose that in a WebID profile a Service Hub can specify which type of services are provided. |
Yes, that last one seems like the preferred option. The URI in the actor is preferably the WebID or some other profile doc of the organisation.
Wouldn't the notification already indicate what the ServiceHub did with the artefact? Not sure whether this is the kind information we should be deriving from the ServiceHub type. Isn't it enough to know that some ServiceHub archived my artefact without knowing what kind of servicehub it is? |
todo:
|
A Service Hub can have multiple roles actually. In Mellon context:
A Service Hub can advertize these roles in a WebId profile:
But it can also specify the role it takes in the Activity:
|
So here is something we need to hash out: are Orchestrators, collectors and datapods? I think not, because that means they would all need to implement server requirements (aka LDN Receiver), which is too much . Can they be servers? Sure. But that is not crucial for the interoperability aspect and thus not part of the spec. But can they have an inbox? Sure, as long as they don't host them (aka LDN Consumer), because that would mean they are servers. |
Ok, I miscommunicated this a little. What I meant we maybe need a vocabulary of There second thing is that orchestrators, collectors can have inboxes even when they themselves aren't services. An inbox is just a LDP Collection with some ACLs ..these inboxes can be anywhere on the web. If you can talk to the network, then you can have an inbox. |
Okay, then we're on the same page. Something like subclasses of as:Service
subclasses of as:Application
?
Indeed, 'have an inbox' is ambiguous. I find the distinction clearer when using the LDN roles sender, receiver and consumer. |
|
yep yep! What I would say is: daen = decentralized artefact exchange network aka this generic layer we are building subclasses of as:Service
subclasses of as:Application
Mellon/ResearcherPod aka scholary communication use-case subclasses of daen:ServiceHub
subclasses of daen:Browser
ErfgoedPod aka Cultural Heritage use-case subclasses of daen:ServiceHub
subclasses of daen:Browser
|
When sending / receiving activities out of the blue, the Orchestrator needs to guess from the
actor
type (or possibleorigin
type in case of COAR) which type of service is contacting the pod. An activity from a review service might need an other rulebook than an activity from a discovery service.Currently all the orchestrator gets is:
Some Organization is contacting me
or using a COAR example
Some Service is contacting me
Wouldn't it be better to have something like?
The text was updated successfully, but these errors were encountered: