OK _ it would be great to try to frame this as a User story if possible please. Should be straightforward and gets us to the focus of what you are trying to achieved and for whom. This helps us make judgments about what we have achieved - it's the target!

Is this the User Story? You know best! As a visualization infrastructure developer and manager I need to reliably map multiple data sets with different characteristics to appropriate visualization types So that I can efficiently and effectively select and publish appropriate visualizations from data sets with known structure but unknown content

Please add one! It really helps frame the problem and gives you a point of reference for the reflection as well as helping communicate the problem that you are trying to solve while letting readers know whether or not they have similar roles and problems that might make this relevant.

OK _ links to SCRC data sources would be helpful.

I think this is OK, but do consider framing this in a way that makes : - the 'data', to which you apply the idiom, the metadata about the data sets and the tacit knowledge that you have about which kinds of plots might work. I think this might be the data that you actually apply your solution to, and you link these pieces of data in your ontology. Worth considering?

OK - so this is a new method that you are presenting in the 2022 paper. Nice.

Nice - very clear - can we get an image or two (maybe from the paper) that shows how this works? Can you explain the methods very succinctly here?

Helpful! What do you mean by "new visualizations"? New instances of existing visualizations (new data)? New visualizations to the system (as in, expanding its range from what might be available), New idioms (unknown to the World until you generate them). It would be great if you could clarify.

I'd probably try to remove some of this stuff from what will be a public facing document. Good that you had fun - can you generalise a little ;)

The first comment is reasonable critical reflection on Tableau. But that's not the idiom. Is there something to say about the idiom that is the focus here?

This is great - very interesting and something we can relate to other experiences.

EF - that's great - almost an idiom in itself: visual scanning for visualization propagation quality check. I hope you write it up!

Were there any things that didn't work in terms of the ontology ... or is it such a generic thing that you can fix it by updating connections? Were all data sets adequately described fo example?

This is really nice and evidently there is lots in the paper! Can you add some figures and examples to show how it works? This of this as a coffee table book that explains what you have done to a broad audience so that other scientists in other domains might be able to understand how it applies to them.

There’s lots of useful information in here, but responses to the original comments and feedback are a little limited. The notebook needs to be completed and shared up to comply with the format we are trying to use. I make plenty of suggestions that may help, or may prompt good solutions. We need this one - please get it into shape! Possible User Story? You need to write one! - As a visualization infrastructure developer and manager - I need to reliably map multiple data sets with different characteristics to appropriate visualization types - So that I can efficiently and effectively select and publish appropriate visualizations from data sets with known structure but unknown content At the moment this is framed as the introduction of something new. That’s not really the way we intended doing this. You can talk about new things in this model, but as developments of old things. We need to show the innovation. E.g., given the above user story, you could take Mackinlay’s model as the idiom. You explain it (3), show where it has been used in the literature (4),show where it has been used in practice (5), before explaining how you tried to apply it, or bits of it, and how you adapted and developed it in the VIS context … which might involve having to really do something new, like the thing you write up in the 2022 TVCG paper, which you can briefly introduce here without precluding that! Maybe that’s too much. Maybe it’s better to use the Khan et al (2020) idea as the idiom, I leave it to you. But do check the comments to try to communicate what you have done in ways that are in line with other notebooks. They should help. Make sure authors are fully described in spreadsheet and notebook and consistent between the two!

Great - and did the idiom (which I think is _the propagation of map multiple data sets with different characteristics to appropriate visualization types_) enable and allow this by, for example, giving you candidate visualizations to plug newly specified data types in to? Which you must have tested to see whether they worked? And this was probably informed by visualization knowledge and literature? It would be great to know how this worked as the tech is great and the claims are really impressive too.