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
Issues #26 and #6 depend on some concept of tracking historical data. What approach should we take to doing this?
Currently:
we broadcast individual network messages upon receipt
WiFi nodes post generated and received messages to the API individually upon generation/receipt
Logging nodes write message data individually upon generation
There is no concept of TX ACK. Ideally, bounced (network re-transmissions) messages would be read as an acknowledgement of network receipt of outbound messages
We want to consider what it might mean to batch data where possible and also to track whether individual messages have been received by adjacent nodes. A likely historical structure might be a simple array of structs. Something like:
{
Message * msg;
bool ack;
}
But will this be sufficient for batching and tracking? Consider sketching a functional API of how this historical structure will be used by code.
Also consider how this might affect the current concept of max ID for a given node. Should this be tracked within this structure? Is there a better way to keep track of message ID numbers?
The text was updated successfully, but these errors were encountered:
Issues #26 and #6 depend on some concept of tracking historical data. What approach should we take to doing this?
Currently:
We want to consider what it might mean to batch data where possible and also to track whether individual messages have been received by adjacent nodes. A likely historical structure might be a simple array of structs. Something like:
But will this be sufficient for batching and tracking? Consider sketching a functional API of how this historical structure will be used by code.
Also consider how this might affect the current concept of max ID for a given node. Should this be tracked within this structure? Is there a better way to keep track of message ID numbers?
The text was updated successfully, but these errors were encountered: