Initial draft of the array APIs SciPy proceedings paper

[asmeurer]

I have an initial draft of the array APIs SciPy proceedings paper here.

There are still some TODOs in the document, most notably I still need to come up with a good motivating example to show in the introduction, and potentially to make reference to throughout the document (see #2). Any suggestions there are welcome.

I'll be on PTO for the next week, so if people want to take a look, please do. Feel free to comment here any suggestions, or to push or PR suggestions as well. I'm particularly looking for high level suggestions, like if the overall outline of the content looks good, and if there is anything that I omitted or anything that should be trimmed down. A couple of pieces of the text are cribbed from the spec.

Here's a PDF to give an idea of what the final thing will look like. There is an 8 page limit (not including references). (I'm not going to keep this updated though so please make reference to the rst file)

The deadline for the draft submission is May 26. Anyone who wants to be any author on the paper will need to sign off on it, and also will need to be added on the SciPy website.

[asmeurer]

By the way, to build the PDF:

• Clone https://github.com/scipy-conference/scipy_proceedings
• Create a new directory papers/01_arrayapi/ and copy the paper materials (paper.rst and the diagrams like dtype_promotion_lattice.pdf into that directory). Alternatively you should be able to just symlink or clone this repo to that directory.
• Run ./make_paper.sh papers/01_arrayapi/ and open output/01_arrayapi/paper.pdf.

One unfortunate thing I noticed: they aren't using Sphinx, just plain docutils. So I don't think it it's possible to add captions to the code examples, as that's a Sphinx only feature. If we decide we want that we will need to work with the proceedings team to figure out how to do it.

[kgryte]

If it's not too much of a lift, may be nice to setup a GitHub workflow which automatically builds the PDF on each commit, as part of CI.

[asmeurer]

I have set up a GitHub Actions build which saves the paper as an artifact. You can download the artifact by going to the latest build at https://github.com/data-apis/scipy-2023-presentation/actions and clicking the artifact at the bottom. Unfortunately, the artifact is always saved as a zip file. If that's too annoying, we can instead add something that uploads it to a branch on the repo using https://github.com/marketplace/actions/deploy-to-github-pages

[kgryte]

Thanks for adding the workflow!

[leofang]

btw @asmeurer do we have the latest adoption/coverage data? It'd be nice to refer to in the paper, I think.

[kgryte]

Adoption and coverage is still not great, so we'll need to think about how to weave the narrative.

[leofang]

I see, so maybe just refer to the data that we collect & tool that we use for tracking the adoption?

[kgryte]

Agreed. Will add.

[asmeurer]

I have submitted the paper https://github.com/scipy-conference/scipy_proceedings/pull/822. If you want to make any further edits to the paper, please make sure those edits get added to that pull request.