Explicitly annotate non-person etymology

[eginhard]

Fixes EqualStreetNames/equalstreetnames#618. Setting gender to '-' instead of null when there is an etymology, but person==False, allows to render it differently from completely missing etymology (here just a subtle difference in line width).

See these screenshots of one street with person etymology, one with non-person etymology, one without etymology.

NB: I also changed the dark mode colours for this screenshot because the current unknown colour is almost the same as the street colour in the map layer, not allowing to see differences in line width: eginhard/esn-module-website@78df489

[eginhard]

@jbelien What do you think about this?

[jbelien]

@jbelien What do you think about this?

I didn't have time to look into yet ; I'll try to do that in the next few days.
Looks promising 👍

[eginhard]

@jbelien Just a friendly ping :) Let me know if there is any way I can help to move this forward!

[tacsipacsi]

Thanks for the PR, it’s good to know I’m not the only one annoyed by this. ☺ In EqualStreetNames/equalstreetnames#787, I proposed using color (black in light mode, could be white in dark mode) – I haven’t noticed the difference in width –, and I still think it may be more obvious than the width change alone. However, given that the colors are defined in https://github.com/EqualStreetNames/module-website/ and thus need a separate PR anyway, maybe give the width alone a try.

[jbelien]

Sorry last year has been pretty busy and I didn't have time to really look into this.
It's now waiting for a year and a half, my apologies for this long wait.

[jbelien]

@eginhard Could you rebase your PR on the master branch? Thanks.

At first sight, I'm not such a big fan of that - but we indeed need a way to make the distinction between no information (false) and information but not a person. Let's then go for this -!

@tacsipacsi Indeed feel free to open a PR in https://github.com/EqualStreetNames/module-website/ to change the layout of the lines having information but not a person based on that -.

Thanks to both of you for your proposal and your patience.

[tacsipacsi]

At first sight, I'm not such a big fan of that - but we indeed need a way to make the distinction between no information (false) and information but not a person. Let's then go for this -!

Actually, - already means non-person, see https://github.com/EqualStreetNames/module-process/blob/master/Command/StatisticsCommand.php#L128, so this PR just uses the existing convention.

@tacsipacsi Indeed feel free to open a PR in https://github.com/EqualStreetNames/module-website/ to change the layout of the lines having information but not a person based on that -.

Thanks for the encouragement, but by “give the width alone a try”, I meant having this PR merged, wait until the next update, see the result in action, and open the PR over there only if the result is insufficient. (And if the result is indeed insufficient, the next question is whether I understand the code enough to open a PR.)

[eginhard]

Thanks you @jbelien! Just rebased the PR.

At first sight, I'm not such a big fan of that - but we indeed need a way to make the distinction between no information (false) and information but not a person. Let's then go for this -!

Actually, - already means non-person, see https://github.com/EqualStreetNames/module-process/blob/master/Command/StatisticsCommand.php#L128, so this PR just uses the existing convention.

Yes, I just followed that.

[jbelien]

Actually, - already means non-person, see https://github.com/EqualStreetNames/module-process/blob/master/Command/StatisticsCommand.php#L128, so this PR just uses the existing convention.

Right, I completely forgot about that (code written 4 years ago) ... My apologies.

[jbelien]

You can see the result here: https://equalstreetnames.brussels/

[eginhard]

Thanks a lot for merging! Glad to see this live! :)

[tacsipacsi]

Thanks! It looks good in Brussels when I zoom in, which solves 90% of the problem, although the difference becomes invisible as I zoom out, so I can’t get a “big picture”. However, I’m not sure whether the color would be better. Maybe I’ll give it a try.

[jbelien]

The goal here is too highlight gender (in)balance ; not Wikidata etymology reference.

It's definitely nice to have and to display that information (obviously) but the focus must remain on the gender information.

[tacsipacsi]

I know, but I believe this effort advances that goal for both editors and non-editor users: by highlighting what the “gray areas” are, i.e. streets about which it’s not yet known if they are named after people (if there are many such streets, the statistics likely don’t reflect reality), differentiating them from streets that are known not to matter for us. As a non-editor user, I’m mainly interested in the big picture, which is the zoomed-out state. (As an editor, I want to reach 100% coverage neighbourhood by neighbourhood, so the zoomed-in state is more important – which already works fine thanks to the current PR.)

If you prefer, I’m also okay with highlighting unknown streets rather than known non-human streets, but I think gray is actually the best color to describe unknown.