Would be good to give a basic list either here or in the above where you introduce metadata, or even a filled-out dummy form that shows some basic metadata. Would help folks who still have trouble visualizing the boundary between metadata and paradata.

Storage and Preservation

Is this what header 2 looks like? Would be good to format these items so they stand out. The text is really understandable and helpful though.

Here introduce the key concepts you define below by saying something like: in order to get as close to a perfectly usable and preservable dataset, you should collect both metadata and paradata.

Before this sentence, introduce paradata somehow, like: "What you can do is record paradata, or information about how and why the data was created, like a data diary."

This is a bit overwhelming - are you saying you should not aim to collect "all the data there are" and set appropriate boundaries, or that you should not expect to avoid all possible hiccups (a "perfect" dataset)? Or both? I think since you used the previous paragraph to point out some specific concerns to address, you should explicitly reference them here. I think you're saying that the key to good datasets is having reasonable and attainable goals, rather than striving to get "all the data and complete in every way possible".

This section is starting to get a bit hard to understand because the data are taking on person qualities. Could you introduce an example of an amusing fake-name project so that folks who do not discuss data in the abstract could more easily follow along?

:

"It is important to plan the structure of the data, expectations for its use by researchers, and arrange for its storage from the beginning of your research."

"Creating good data requires planning"

"And maintaining data in this condition helps protect them from the ravages of time" maybe?

Reword this so it is clear that it is different from the above: Access part = text and numbers and database saved in a bad format but still usable // Reuse part = a map or output saved as an image but the underlying data used to create it is not present (did I understand correctly?)

Make sure somewhere earlier in the text someone has defined "proprietary" in plain language

?

Remove this question - the following section needs to be read first otherwise it is overwhelming

remove "in the past" and change following to "when trying to use someone else's data or reconstruct their conclusions"

I assume this is going to be turned into a link to a previous chapter?

Remove