JOSS paper

[JulienPeloton]

Following https://joss.readthedocs.io/en/latest/submitting.html, we can open a branch and push the paper.

[Christovis]

Hi Julien, a JOSS reviewer here.
I read your well written paper and was just wondering whether I might have overlooked a part that describes how s4cmb "compares to other commonly-used packages in this research area". If there is nothing to compare to at the time of publication, it might be informative for a user to know that.

[JulienPeloton]

Hi @Christovis

Thanks for the feedback. As far as we know (including @gfabbian ), there is no alternative to s4cmb if one asks simultaneously:

• publicly available code
• Generic in the type of simulated effect and experiment
• self-contained end-to-end simulations (sky -> project to timestream -> inject perturbation effects -> project back to sky)

In the field of CMB, the codes to simulate and study systematic effects are usually:

• publicly available (and described in a research paper) but focusing on one specific effect, and often in the map domain only (no timestream generation), or
• internal to a collaboration and not publicly available (e.g. Polarbear, SPT, ACT, Simons Observatory), and/or
• part of a huge data analysis pipeline that makes them difficult to use in a different context (e.g. Planck - or any satellite data processing).

There is one notable exception, TOAST which is a general purpose framework to simulate and analyze CMB observations (Disclaimer: we know and collaborate with the authors). Even though one could potentially extend it to support a wide range of systematic effects, this is not their focus (rather efficient TOD manipulation on massively parallel architectures).

I've clarified the text in the paper (pushed).
As far as we know, s4cmb is the only dedicated package that enables the study of a wide range of instrumental simulations, from the instrument to the sky map, while being publicly available. For more general purposes, including some instrumental systematic effect simulations, users might also consider the use of TOAST [ref], a software framework to simulate and process timestream data collected by telescopes focusing on efficient TOD manipulation on massively parallel architectures.

ref: https://github.com/hpc4cmb/toast/

[Christovis]

One more thing. The README could contain a clearer "Statement of need" in your "The package" section.
I think the "Statement of need" in the paper is quite good, so that I would suggest to add a shortened version to the README (e.g. leaving out parts such as package dependencies and containerizing as you mention them later on...).

[JulienPeloton]

Thanks for the suggestion!