#JavaScript

The test results in #18 demonstrated the bug for many years, until patch testing got turned off.

I'm not disputing that, just trying to help make it easier for anyone interested in picking this up. Without this it's hard to know at a glance if anything has changed in core that might have fixed this over the last 5 years.

Thought about this again, and still think 'unique aggregate' will work. Per #18 don't think this is postponed on anything.

This needs an issue summary update, let's start with the standard issue template. I think it's also not clear from the title what is the bug that needs fixing, i.e. the is-active class is not added. Let's get 2525830-18-комацу.patch converted to an MR so we can see the test fail, then we can work on a fix for this.

#1. I've updated the statcounter link to go directly to https://gs.statcounter.com/browser-market-share, which includes regional filters by continent and country along with a short mention. I think it's good to link to the resources we know about, but don't think we want to be comprehensive/prescriptive here in case something better shows up.

#2. I had a look at the most recent webaim survey to see if there was an example.

https://webaim.org/projects/screenreadersurvey10/#browsers

This shows Firefox at 16% and Edge at 19%, whereas if we look at the last month's global data on statcounter:

https://gs.statcounter.com/browser-market-share/desktop-mobile/worldwide...

Firefox is at 2.5% and Edge is at 5%

So I've added "(e.g. more than 10-15% stable usage when global usage is under 1%)".

The 'stable' in there is so that the 'downward trend' above can take precedence, this was part of the considerations for dropping IE11 support iirc because it was clearly on its way out in successive webaim surveys, if still disproportionately used.

Stable 10-15% usage isn't a hard number (and we could change it to a different number), it just means we would need to put some effort into understanding why a browser is disproportionately and consistently used over the years before dropping support for it.

#3. I started writing a comment saying I think versions is already covered in https://www.drupal.org/docs/getting-started/system-requirements/browser-... → , but then I realised that doesn't cover how we decide which browsers get two major vs. only latest major version covered. So yes, we probably do need something.

I've added:

In general, desktop browsers are supported for their two most recent major releases, and mobile browsers are supported only for their most recent release. See https://www.drupal.org/docs/getting-started/system-requirements/browser-... → for details of current support.

I think we'd need a good reason to deviate from that for a specific browser, firefox ESR and safari mobile are the only two deviations we have, not sure we can get around discussing special cases like that.

I think that hopefully addresses the feedback, going to be bold and self-RTBC the changes since they are pretty minimal in the scheme of things.

I think this looks great. Few comments that would be great to address to make sure that it's clear how this guidance should be applied:

1. Can we add a reference for geographical browser usage? If we don't have one, I don't think we should include it in the heuristics.
2. Can we define specific thresholds for when we would we consider a browser is disproportionally used by screenreaders per webaim screen reader survey?
3. Should we add guidance on how browser versions are handled? If they follow the existing policy, we could drop support for previous major of Safari for example.