DataTF

From Auto

Data task force

In the W3C Automotive WebPlatform Business Group there is a task force exploring various issues and concerns for application use cases within the vehicle and in off-boarding.

Harjot Singh (Geotab) created a Data Contract document to outline some of the issues from their experiences we should take into consideration. In some cases the data model itself could be improved to address these concerns, in others in belongs more in the Gen2 service specification or could be represented in policy language the task force is exploring if not elsewhere. Below is a chart to try to succinctly list these topics and thoughts on where they may be addressed. Further descriptions can be found in the aforementioned document and can be discussed at the upcoming F2F meeting in May 2019.

Issue description Potential scope Comments
accuracy of a given signal, if known VSS related: error thresholds
sampling methodology VSS or separate ontology like PROV event, timed, curve etc. primary use case is to add value to data in-cloud but also useful in-vehicle
signal access Gen2, policy, guidelines policy from OEM app marketplace
pending signals VSS (eg NULL, 0 or -1) or Gen2 results plus error code how to communicate to app one or more signal it is entitled to and subscribing or retrieving is not ready yet
signal duration (ttl equiv) VSS duration a signal should be considered good before repolling or expect a fresh one if subscribed. similar to sampling methodology
transmission use Example calculate costs, be more efficient. was deemed earlier by tf to be out of scope, JLR expects TCU to be proprietary, Ford interested in serialization/compression transmission standardization
in-vehicle storage guidelines and policy whether and how much local storage an app can have
personal identifying information guidelines, policy, VSS we may want to mark some signals as clearly PII eg VIN, but combinations of signals could be used to fingerprint a vehicle or driver and evolve over time
Example Example Example
Example Example Example
Example Example Example