Technical paper - use case description for SI

[anerv]

No description provided.

[anastassiavybornova]

👀

[anastassiavybornova]

@anerv a (very!) first draft of the SI is ready for you to check. plots are more like place holders for now, using them just to illustrate what my general idea of the SI would be - i.e., explaining for each layer/evaluation step (ordered in the same way as Table 1) how the information provided by the BikeNodePlanner can be helpful in revising the network. happy to get your thoughts on the general structure, and then figure out together how to distribute the polishing of each SI subsection.

[anastassiavybornova]

SI todos:

• move SI to separate .tex file, add authors, affiliations etc.
• add DKNT references in SI @anerv
• add motivation (couple of sentences) for each subsection - based on handbog? @anerv
• play around with seaborn (and others?) to get a better palette for evaluation layers (ideally, where nature "happens" to be green, etc) (picking happens in

  bike-node-planner/scripts/script02.py

  Line 108 in 12032bb

  layercolors = sns.color_palette("colorblind", len(layernames))

  )
• check disconnected components in Fyn - why? @anastassiavybornova
• Add legends to SI figures
• reference SI-figures from main document
• Update figures in SI with new colors

@anerv thanks for the SI revision! little to do list above, i added our names where it made sense for me, and the rest is up for grabs, i hope to finish this by the end of the week so we can show it to the others :)

[anerv]

• Should each figure in SI be referenced from the main file?

[anastassiavybornova]

Should each figure in SI be referenced from the main file?

i think since it's a short paper, the current solution (manually writing out the reference when we use smth from main in si) is good enough, yes! (like: "Table 1 from the main text". or maybe i misunderstood what you mean)

[anerv]

No what I meant was more that usually there should be a in text reference to all figures - so whether we should enforce that or not :) On 20 Nov 2024, at 11.46, anastassiavybornova ***@***.***> wrote:  Should each figure in SI be referenced from the main file? i think since it's a short paper, the current solution (manually writing out the reference when we use smth from main in si) is good enough, yes! (like: "Table 1 from the main text". or maybe i misunderstood what you mean) — Reply to this email directly, view it on GitHub<#51 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ALNM6P6C5O6QIZUQKC45TFL2BS4E5AVCNFSM6AAAAABR3MS3E6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIOBZGA4DOOBUGQ>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[anerv]

@anastassiavybornova for the color selection, isn't it okay that the colors are random if you do not provide automatic colors? But we would provide a commented out suggestion in config-colors-eval matching the default categories - and then use our own suggested colors for the plots in the paper?

I've added a list of suggested colors in the config-colors-eval (see pull request) using colors from this site: https://personal.sron.nl/~pault/

Feel free to request another solution!

[anerv]

@anastassiavybornova let me know if you're okay with the suggested colors for the evaluation layers - then I'll update the plots in the SI and main document

[anastassiavybornova]

Hi! @anerv sorry for the late reply on this!! Yes, that's great with the color suggestion & the plot updating :)

[anastassiavybornova]

oh and:

usually there should be a in text reference to all figures - so whether we should enforce that or not :)

i would say yes!

[anastassiavybornova]

@anerv

check disconnected components in Fyn - why? @anastassiavybornova

resolved by #66 :)

i've also updated the disconnected component plot in the overleaf, but i think it needs another revision round from you (limiting the study area extent for figure 2 in main, and adding the legend for the fig in SI? for both, i'm not sure how you went about it, but let me know and i can also take care of it!)

[anerv]

I can take care of the SI - it is just manual clipping together screenshots so nothing fancy there.
For the automatically generated plots in main, it looks like it is zoomed to the study area/the largest layer, but the data used for the components map also include Langeland (unlike the others). So it's more a question of streamlining which municipalities are included so it's the same for all plots.

I'll rerun data generation + the node planner over the weekend and update all plots accordingly, including new colors. Any preferences for whether to include Langeland or not?

[anastassiavybornova]

@anerv huge thanks!! i would include Langeland, yes (see #24 -- list of municipalities agreed on with Kirsten for the Fyn region)