You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
There are two potential cases related to subscription matching and broadcasting to take into account, which consider broadcasting of past events, which are suddenly matching because of some new event that has arrived on the stack. Examples:
Zenodo subscribes to the relation events where the "target" is a Zenodo DOI. Let's say the following is on the broker stack:
and the following event arrives: 2010YZCDF..84..200F cites 2004MNRAS..84..308E (by ADS)
At this point from the last event itself it's not clear that Zenodo should receive the information, however if the broker was to resolve it's own stack to see if the "target" bibcode 2004MNRAS..84..308E is the same as Zenodo DOI.
A more complicated example:
2.
Let us assume the opposite and let's say that the following is on the broker stack:
At this point it is clear from the event that Zenodo should be informed, however, broker could also go back in history and see if any of the past events which were not sent to Zenodo, should with the new information about the identity between identifiers.
Potential issues:
Finding out a single identity relation, could spawn a large number of past events to be broadcasted.
Broker should somehow keep notion of what of the past events were not sent to Zenodo.
The text was updated successfully, but these errors were encountered:
krzysztof
changed the title
broker subscription and broadcasting history usecases
broker subscription and broadcasting from the past usecases
Oct 12, 2017
There are two potential cases related to subscription matching and broadcasting to take into account, which consider broadcasting of past events, which are suddenly matching because of some new event that has arrived on the stack. Examples:
Zenodo subscribes to the relation events where the "target" is a Zenodo DOI. Let's say the following is on the broker stack:
and the following event arrives:
2010YZCDF..84..200F cites 2004MNRAS..84..308E (by ADS)
At this point from the last event itself it's not clear that Zenodo should receive the information, however if the broker was to resolve it's own stack to see if the "target" bibcode
2004MNRAS..84..308E
is the same as Zenodo DOI.A more complicated example:
2.
Let us assume the opposite and let's say that the following is on the broker stack:
Then the following event arrives:
At this point it is clear from the event that Zenodo should be informed, however, broker could also go back in history and see if any of the past events which were not sent to Zenodo, should with the new information about the identity between identifiers.
Potential issues:
The text was updated successfully, but these errors were encountered: