Define scope

[EmmaWoolliams]

We need a discussion about the scope. In your opening you talk about manufacturing and health, but not energy, transport, high-end physics, biodiversity, etc etc.

If you do want to cover all branches of physics, then we need to bring more people in. If you just want to say that these approaches have broader applicability, there are probably better ways of doing that.

I think it's also unfair to say all these communities "often overlook" "uncertainty and error correlation information" - especially as they almost all do consider uncertainty information at least.

In my paper I was writing this (which might give ideas):

This approach to metrological traceability has supported global industry and trade, and cutting-edge laboratory research for 150 years, providing confidence in measurements'. More recently, with the accelerating digital revolution, metrology institutes have begun to look beyond confidence in or traceability of measurements' to confidence in or traceability of data' and confidence in or traceability of information'.
Such efforts have also come out of reactions to the challenges of obtaining reproducible measurement results \citep[e.g.,][]{milton2020,plant2018}.
In moving out to data and information, metrological traceability moves beyond a `chain of calibrations' to a traceability chain that includes both calibration and data-processing steps.

[EmmaWoolliams]

Overall, I think we need a new introduction, but also we need to think about the scope separately.

My suggestion for a new introduction is something along these lines:

For users to assess the fitness-for-purpose of a dataset for their application, it is important that datasets have associated with them sufficient information about uncertainties. Today, measurements are ubiquitous in (for example) science, manufacturing, trade, health, transport, energy and in monitoring the environment. In many of these applications, datasets are getting larger, with many interconnected measurements, for example providing timeseries, spatial information or from sensor networks. For such datasets it is not sufficient to consider only the uncertainties, but also to consider (error) covariance structures between different individual measurements. At present there is no clear example of a data and metadata standard to collate information on such error covariance structures. ...