Update flow chart

[bnord]

Current flowchart Created in google slides

• [x] we should update the figure after we've finalized the variable names. e.g., PS_maker is not a great variable name. see issue #69
• [x] noise tag variables are hard to understand what they mean
• [x] variables in one of hte boxes are on different lines in the box; should be on same line
• [x] image should be at higher resolution. the png looks pretty low res even when opened by itself on a web page.

[bnord]

1. I think each module box should have both text and a symbol, and these terms should be directly related to what's in the code.
2. I think we should consider this as part of what would go into the software documentation.

[cavestruz]

• [ ] add to the readme for front visual for new users.

[samueldmcdermott]

thoughts on this new one? ex_workflow_alt.pdf

[samueldmcdermott]

(also: assigned this to @humnaawan )

[bnord]

• [x] Cosmology (r, Alens): "arrays" is unnecessary and takes up space and is confusing.
• [x] To be with the above in parallel, there should be anObservational category similar to Cosmology. then the properties should be turned into parameters in the figure, like the r and Alens for cosmology.

[bnord]

• [x] where does the CAMB config get called read in?

[bnord]

is renaming warranted?

?rename: params_io --> parameters_input_output? ?rename: config_obj --> configuration_object?

[bnord]

• [x] "output" should be clear through the geometry of the box and the rest of the figure. it shouldn't need to be explicitly labeled.

  

• [x] the labeling should be more concise: calling it a dictionary is unnecessary.

[bnord]

• [x] could be power spectrum coefficients

[bnord]

Does the plotting module still exist? If not, we should still state that we use namaster if we keep that in there.

[samueldmcdermott]

is renaming warranted?

?rename: params_io --> parameters_input_output? ?rename: config_obj --> configuration_object?

if that is a question, the answer is I don't think so (these names were suggested by others and I think they're clear), but if that is a direction then I am happy to take it. Please let me know if you do in fact have a preference for these changes

[samueldmcdermott]

Does the plotting module still exist? If not, we should still state that we use namaster if we keep that in there.

yes -- it is still there but it is optional (an "extra" in the parlance of the pyproject.toml). I was planning to specify in the caption or text that that is optional (hence the dashed line). It already states that it calls namaster -- are there other changes you would like on this, or is it okay?

[samueldmcdermott]

@bnord I edited most of your comments to include checklists, which I checked, but two of them required clarification, which I added in replies above. Let me know if my questions make sense, happy to go forward in any way with them.

New figure here (low-res for now but will improve that before putting into the paper)

[samueldmcdermott]

@bnord tagging you again

[bnord]

is renaming warranted? ?rename: params_io --> parameters_input_output? ?rename: config_obj --> configuration_object?

if that is a question, the answer is I don't think so (these names were suggested by others and I think they're clear), but if that is a direction then I am happy to take it. Please let me know if you do in fact have a preference for these changes

If we remove matplotlib, should we remove the cl_plotting function from the flowchart? I'm thinking yes if I'm caught up on everything.

[bnord]

oh, sorry, I just saw that optional is a possibility. I would guess that it would be best practice for clarity to indicate the cl_plotting is optional in the figure.

[bnord]

I think I'd like to do some more tweaks to the figure, but it's probably inefficient (and maybe annoying for others) for me to make requests little-by-little. Do you want to re-assign this to me, and I can make a try?

[samueldmcdermott]

just assigned it to you @bnord -- feel free to make any edits you like in the google doc (or any way you like)

[bnord]

How does the latest version look?

https://docs.google.com/presentation/d/1-gDIkdLtom0yEvgFSj0C8N1egvZOjfOjp9rjqfLfI6U/edit#slide=id.g24eb1e1981e_0_60

[bnord]

here's my logic:

1. solid blocks are for files, which could be scripts or a data file
2. dashed blocks are for data elements

I know this doesn't adhere to the schools of thought for flowchart design, but I think it might be sufficiently self-consistent and human-readable.

[samueldmcdermott]

looks good to me! Only Q -- did you mean to say max_l twice in the lowest block? In CAMB there are both max_l and max_l_tensor so maybe the second one was supposed to add _tensor to it?

[bnord]

I updated it. Does that work?

https://docs.google.com/presentation/d/1-gDIkdLtom0yEvgFSj0C8N1egvZOjfOjp9rjqfLfI6U/edit#slide=id.g24eb1e1981e_0_60

[samueldmcdermott]

looks good to me

@humnaawan do you want to close this? If not done by, say, 2PM ET I'll close it