Consider adding Du Bois Data

[ajstarks]

I've recently learned of this repository, and I'm wondering if you will consider adding data from the Du Bois visuals created for the 1900 Paris Expostion. (I've collected such data in my repo at https://github.com/ajstarks/dubois-data-portraits

[friendly]

Yes, I would be thrilled to have some of these datasets in the HistData package. I've followed your work and looked at some of your decksh repos. But now I'll have another look at the new repo.

[ajstarks]

Great. Where feasible, there is a "data.csv" for every plate. For example: https://github.com/ajstarks/dubois-data-portraits/blob/master/plate27/data.csv is the data for the "fanchart".

…

On Thu, Mar 16, 2023 at 2:02 PM Michael Friendly ***@***.***> wrote: Yes, I would be thrilled to have some of these datasets in the HistData package. I've followed your work and looked at some of your decksh repos. But now I'll have another look at the new repo. — Reply to this email directly, view it on GitHub <#3 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AABJNXGH7DORB6NRAQAEPXLW4NIULANCNFSM6AAAAAAV5RTO3I> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[friendly]

OK, I had a look. There is indeed a separate data.csv file for each plate, and this would not work within the structure of the HistData package, where the primary emphasis is on more comprehensive data sets that can be approached in different ways. In a few cases, e.g., Snow's data on cholera, I had to create Snow.deaths, Snow.pumps, Snow..., but generally these are for a single kind of analysis / graphic.

There is certainly merit in making DuBois' data available in an R package, to encourage wider exploration of his graphics.
I'll begin to think about how this could be done, and what a DuBos package might look like.

[friendly]

I have no problem importing CSV files. The problem is that there is a different data file for each graphic, so it would make more sense to create a separate DuBois package. I'm willing to work on this if you are willing to lend a hand.

First, I would need a relatively complete set of the data files, cued to the DuBois catalog. I think Anthony Stark has these for the DuBois challenge.