So a big issue is this drops support for anyone using Windows 7 and Windows 8.
As the max versions of Firefox there are 115 and Chrome is 110.
This is about 800 of my active users.
And I am not sure about mobile OS’s this affects?? But probably many more. Half of my users are not in the US and I thinks this disproportionately affects them.
has bigger problems than not being able to use Discourse.
I’ll repeat Firefox’s statement here as it is a sensible summary of the situation:
Microsoft ended official support for Windows 7, 8 and 8.1 in January of 2023. Unsupported operating systems receive no security updates and have known exploits. With no official support from Microsoft, maintaining Firefox for obsolete operating systems becomes costly for Mozilla and dangerous for users.
This was 2+ years ago.
There’s options. Get a new computer. Get a chromebook. Get a phone. Use Linux.
I’m not sure what you’re trying to imply with this statement, but most people are not in the US. Most of our staff are not in the US.
I fully understand and agree with the security implications of maintaining support for obsolete operating systems. However, my concern here isn’t about Microsoft’s end-of-life decision or whether users “should” upgrade, it’s about real-world impact on communities currently using Discourse.
In my case, this affects ~800 active users, many of whom are in regions where upgrading hardware or OS is not trivial due to economic or logistical constraints. The assumption that everyone can “just get a new computer or a Chromebook” doesn’t hold universally, especially outside North America and Western Europe.
The disproportionate effect on international users is not a rhetorical point, it’s a visibility concern. If our mission includes enabling global discourse, we should at least acknowledge the cost of cutting off access for a segment of users who may rely on this software more than most.
I’m not pushing to reverse the decision, just advocating that we approach it with awareness and perhaps provide clearer messaging or transitional guidance for those affected.
I understand this problem and want Discourse to reach as many people as possible. That said we also need a modern experience and we need to take a stance around the minimal requirements for a modern experience.
I feel like the long term possible solution here may be a “html+css” writable view.
All your 800 users will still be able to read and interact via email, that is not going to be lost.
Maybe the long term change here is that we will eventually provide a minimal ability to login and write in our html view so some of the more unfortunate can still interact on very old devices.
It is important to note that bargain oppo and techno phones will still run Discourse after this change, we are talking here about hand me down devices generally or people who can not afford to upgrade for 10+ years.
I am on Android 9 Kiwi Browser 132, using Chrome 132.0.6961.0, but that colour feature is still shown as not supported and I have that warning banner.
See my Feature detection result in WebDev console:
{
"relativeColor": false,
"subgrid": true,
"lookbehindRegex": true
}
@david, maybe rename this topic as it’s not limited to some iOS version, but rather to some browser versions on any OS.
But sorry I have no title idea.
Does current Firefox ESR (older versions) pass the test?
Chrome 132 sounds like a recent version.
I don’t know Chrome versions.
Isn’t it too early to enforce the use of relative colours?
Potentially related:
Are you 100% sure that Kiwi is using Chromium 132? It’s strange that they would do that, and not support relative colour syntax. Could you try visiting @Arkshine’s test page and see what it says?
On ‘official’ google chrome 132, it’s definitely supported:
Yes, the current ESR is 128, which supports all these features.
Updated 
The latest Kiwi version is v132: Releases · kiwibrowser/src.next · GitHub
I installed it:
It seems that it should initially be available from v131 for desktop/android/webview, but caniuse says it’s available from v134 on android. However, I can’t find any reliable information about it (probably not looking at the right place)
That might just be because they haven’t collected any data for earlier versions of Chrome on Android. Note there are no red squares above 134 to indicate a lack of support in older versions.
Unfortunately it seems like testing an older version of Chrome on Android is quite tricky. Even browserstack doesn’t do it.
Doesn’t all this just…go without saying? This applies to literally all technology. All technology for anything ever has always been, and always will be, limited to some people. That’s just how it is.
Discourse’ alternative options are stop updating and therefore let their software stagnate, eventually open up critical flaws, and possibly get pushed into irrelevancy by someone not doing this.
Or option B, pay and support an entirely separate team to ensure full support for everything, always, forever (because what’s the cut off? 12 year old phones?). This is wildly unfeasible.
For context Windows 10 currently has 60% of the desktop marketshare, and 11 has 35% (and these are a few months outdated at the very least). This means unsupported (desktop) users come in at a whopping single digit percentage. So that a significant enough portion of people accessing discourse with ZERO way to do so exists in order to require unlimited unending support for everything just isn’t reasonable.
It sucks, of course, but that’s a inherent part of technology
Bu here relative colour syntax sounds gadget compared to how recent Chrome 132 or 134 is.
Continuing the discussion from Dropping iOS 15 & other old browsers in May 2025:
I feel like people using old web browsers should still be able to use a basic gui framework, to comment, login, etc… as long as the web browser is still being worked on.
There are a lot of web browsers that fit this qualification.
Badwolf, qupzilla, epiphany, basilisk-browser, palemoon
other uxp based web browser
etc…
Not everyone likes all this bloated eye candy that the web browsers embrace adding such as chromium based web browsers and firefox quantum and forks of that.
I felt the need to mention this because I got an error today on two different discourse using websites and I really feel like it shouldn’t be dictated what web browsers are discriminated against based on modernization.
Each of the current modern web browsers, firefox, safari, opera, chromium web browsers including edge and google chrome all are infected with unneeded features telemetry and in many cases data collection.
So… that’s how I feel about “modern”
Then again, I don’t expect you to support web browsers that haven’t been updated since before the early 2000s.
I would think you’d ignore those and say your on your own.
But yeah, I absolutely hate what modern web browsers look like. They are too bright and shiny. They look awful. I feel the same way about windows xp and newer as well. But that’s neither here nor there.
Anyways, I guess I am just saying, at least do what I mention in the very first point if possible.
[/details]
Not to detail this topic. But might be an idea to keep an eye on “Discorkie” project.
Though if it is just the bloat of the browser. You could just use the “install app” say in chrome to have a somewhat stripped down version of chrome browser as an ap.
I can appreciate wanting to stick with a oreferred browser/platform. However older OSes and browsers often lack newer features and have other security hole potentiomal due to not supporting newer standards.
Thank you, good to know. Does this view allow to create topics and reply, even if in a rudimentary way like missing some convenient features?
Asking for a friend. 
No, asking because a user in our community has seen the banner and is asking about what will happen.
And does anyone know whether macOS 10.13 users have any chance with any browser?
No. Well, yes by using email — if that is rudimentary way enough 
Problem with that theory is that microsoft edge, google chrome, safari and opera are all proprietary software and yet they are all suppoted web browsers.
I have no doubt firefox even current is more secure than that crap. Same with palemoon and other niche browsers especially. Also, privacy issues are very common on proprietary browsers. So that argument you made is kind of filled with holes. No offense… but yeah it makes me cringe.
Is all that work worth the effort? How many people does this realistically affect? 
Do you have any stats on the percentage of Discourse users that will be affected? If you could put something in to experimental that collects anonymised compatibility stats in the background I’m sure most Discourse admins would enable it to gather some numbers.
Discourse is already showing the banner to some users, and I wonder if there is a Data Explorer way to find out who is getting it. This way admins could a) not just wait until someone complains, and b) even if someone complains they can say with baked data “as far as we are aware, [number] active users out of [number] are affected by this problem.” I’m not requesting the creation of anything, just asking whether this is possible already with a database query.
At the moment, no. We don’t send any telemetry about JS features to the server, so it wouldn’t be available in data explorer.
The best you could do is analyze the User Agent strings in the NGINX logs to look for certain browser versions.
