This link does not lead to the specifiy discussion. Did the renderers agree to or comment on the dimming?
Website changes are discussed on the github repo and it has been that way for more than a decade. Now you might find that good or not, but it doesnāt make sense to pretend that that hasnāt been like that for longer than most people have even been active.
As a result if you want to participate in the discussion you are going to need to keep an eye on the repo in one way or the other.
And yes github has a search function and searching for ādark modeā among other things turns up an issue with 52 comments.
2 Likes
It was discussed for over four years on the Github repo for openstreetmap-website, which is where changes to the OpenStreetMap website are discussed.
4 Likes
Although itās already been said, I too appreciate the dark theme in general. But the theme should only concern the web interface (menus, buttons, windows), not the map!
Or else, by proposing to change the map theme independently of the browser theme.
But here, apart from the interface (which I like), thereās no real dark theme for the map, just a dark veil that reduces readability.
Link to the French-speaking forum where I had just reported the problem, with screenshots.
8 Likes
There is a PR (pull request) open now on GitHub which will remove just the brightness filter (i.e. dimming) in dark mode:
So in dark mode the map will look like it did before:
13 Likes
Donāt think this will be merged as is. This is one of four cartography options on the table (#5328) that optimally the user will be able to choose in a toggle (#5324). This is still under discussion.
1 Like
Letās see. If it were up to me, I would remove this brightness filter asap. It really makes using the OSM website in dark mode quite bad, with no opt-out (short of changing your whole browserās theme).
I think the UI with customization options will be great, but that will take time to find consensus on design and to implement.
14 Likes
Personally I also think that dimming the tiles is the worst option, but itās what a few contributers chose, @Anton_Khorev added in #4712 and the option @Andy_Allan is currently pursuing.
4 Likes
I just want to clarify that Iām not pursuing this option! Let me explain.
Most people know I have a conflict of interest in being both a maintainer of the website and also a cartographer for (at least) two of the featured styles. But I try very hard to be completely fair when it comes to things related to the maps and generally stay of the discussions that directly affect them or when something might affect one style and not another. Even when I was on OWG I stayed out of any conversation about which styles should be featured.
Unsurprisingly, I have my own (very strong) preferences on what we should do about the maps in dark mode. But rather than trying to ram through my own decisions, Iāve (twice now in the last six months) carefully documented all the options that the community has discussed, and explicitly asked for feedback. I would rather that the community considers and discusses the options, and more people understand that they all have pros and cons.
But this approach has come in for a lot of criticism over the last 48 hours, often from people who have a preference for one option but arenāt willing to consider the pros and cons of them all. An alternative approach would have been for me to simply try to force through my own preferences, with no discussion about the pros and cons of any other approach. Iād certainly get criticism for that too, I reckon!
11 Likes
Sorry for the wall of textā¦ TL;DR: users, be kinder and thankful to devs; devs, there needs to be wider involvement of users in the decision making process. Also, where should we actually discuss the matter at hand now?
I see the situation is more clear now, this whole issue is mostly due to miscommunication (in general, Iām not blaming anyone).
Being one of the first who pointed out the problem user-side I understand there is some frustration, but still I invite my fellow mappers to avoid being aggressive towards whoās responsible. Iāve read here and in other places some comments definitely too aggressive for my taste.
On the Internet weāre used to express rage and harsh criticism about interface changes we donāt like, but we usually do that as customers of commercial enterprises. This approach is definitely out of place towards a project offered to us by volunteers with little or no help offered by the rest of the community.
So, everyone, be kinder and donāt take for granted whatās being offered here.
This change was far from perfect but itās clear there were good intentions and hard work behind it, so show some support and donāt be an entitled Karen.
That said, Iād like to ask for some clarifications on how as a community we can work together on this issue, and in general avoid miscommunications between devs and users in the future.
I think most of the problem stems from the fact that OSMās community discussion is often dispersed across several platforms.
This surely has been discussed beforeā¦ somewhere 
Iām fairly new to community discussions in OSM, so itās fully possible Iām just speaking out of ignorance, but my understanding of the situation is that dev-side many were convinced Github discussions were already involving all the relevant stakeholders, while many of those stakeholders were convinced main interface changes would have been discussed or at least announced somewhere else (in this forum or through some sort of push warning).
On the other hand, those stakeholders were not victims being kept in the shadows, they were just blissfully oblivious of the work going on under the hood (me being one of them, in the first place).
Having zero experience in programming I barely had an idea of what Github is, let alone know OSM changes are discussed there. Now I know and Iām not pushing to change it: probably Github is a better suited platform for dev works.
Still, I think some wider involvement is needed on changes of this relevance.
My suggestion is to simply push some notifications to logged users when major changes are being evaluated, pointing out briefly whatās going on and where to discuss it.
This would help transparency, avoid these conflicts, but also raise awareness on the effort being put in (and maybe also get some help with the job).
Now, to get out of the current issue, right now where is it useful for end users to express suggestions on the dark mode interface to devs?
Here, Github, somewhere else? Iāve seen several Github threads on this, is there a main one where to discuss?
Thanks to everyone
7 Likes
All I know is now there is no way to see a nice bright shiny road again. Like no matter what toothpaste one uses, itās always going to be dim grim.
Yes, I tried both light and dark modes in the Chromebook appearance settings.
How is the colour change implemented? Since these are .pngās, is a certain value substracted from something? Is this done by the browser?
This is done by the browser, lowering the brightness of the tile layers by 20 %.
And because it is done by the browser, there are already scripts and styles made by others that change that behaviour.
Whilst this OSM dark theme is only a restyling, better-osm-org also adds additional functionality.
4 Likes
Please remove the shady gray layer on the map. I cant use the map at the moment! Awful! Dark theme is greatā¦but not on the map. Or make at least an option to turn it off or to adjust the brightness. Thank you.
6 Likes
A quick userContent.css definition to undo the greying:
/* openstreetmap.org */
@media (prefers-color-scheme: dark) {
#content #map .leaflet-pane .leaflet-tile-container .leaflet-tile, #content #map .mapkey-table-entry td:first-child > * {
filter: brightness(1.0) !important;
}
}
The #content #map selectors I added to mimic openstreetmap dot orgās DOM structure. It might still catch some other websites, but would make no difference unless they use brightness filters too.
7 Likes
Yes, this is done entirely by the browser. The png files being served do not change.
1 Like
This would be no problem if there was a toggle. It should not have been implemented without the toggle.
This is not directed at you, hlfan - your comments here have been relatively helpful - but causing large usability issues for large numbers of users and then dismissing their complaints because they arenāt software developers (i.e. savvy with how to participate on Github, skilled enough to understand the discussions, and with lots of free time to do so) is extremely bad PR for the project and comes across to a lot of the user base (including people who contribute to the project in other ways, such as with data or even money) as obnoxious gatekeeping.
6 Likes
Neither @Richard nor myself have any particular stake in this matter. You are just blaming the messengers for conveying the facts as they are.
2 Likes
Itās my opinion that comments boiling down to āDonāt complain here, unless you know how/have time to use Githubā are not appropriate to a āHelp and supportā forum, regardless of their motivation.
My comment was intended partly as a general complaint about actions taken outside of this forum (as I think is clear from a level-headed reading of what I said), which I assume is allowed? Though I will also say that some of the comments in this thread do, at best, carelessly reinforce the frequently-encountered dynamic of people closer to the development side of an open source project reacting dismissively to the feedback of the broader user base that itās marketed to.
6 Likes
I should probably have read through Wilhem275ās comment more carefully before wading in. Theirs covers most of the same issues as my complaints, but in whatās probably a more productive tone.
I would only add an emphasis that software developers are not, by a long shot, the only contributors to the project. Many others are contributing to the crowdsourced data that the map at OpenStreetMap.org relies on, as well has helping to spread the word about not only the data but that very web interface, not to mention in some cases contributing funding for its operations. That is, many of us are stakeholders not only in the sense of being users, but also of being contributors comparable in importance to the project as software developers, despite not being able to participate in Github discussions.
Is this the right place to make that point? I think it has to be, unless some equally accessible platform is made obvious to those kinds of contributors.
4 Likes