Hi everyone, the Design System Team (DST) is proposing the following changes to MediaWiki browser support [1]: - Drop support for Internet Explorer 11 (IE 11) - Drop support for all versions of Edge Legacy - Drop support for Opera - Increase Basic (Grade C) support for Chrome and Firefox to versions 49+, Safari and iOS to versions 10+.
What this means: The browsers we’re phasing out won’t be tested for layout rendering anymore. While users on these browsers might and will still be able to read and basically interact with content, they might experience some quirks. This step helps us integrate modern web features more seamlessly.
These changes will unlock the ability to use specific newer browser features that cannot be safely used today without requiring a fallback, notably CSS custom properties (used in upcoming reading customization features like Night Mode) and the <summary> and <details> HTML elements that can be used to replace the checkbox hack. This will reduce the amount of code sent to 99.9% of users and cut down on software development costs and maintenance burdens.
See the full announcement for more details; PDF to download [2].
On behalf of the Wikimedia Foundation Design System Team, Volker
[1] https://www.mediawiki.org/wiki/Compatibility#Browser_support_matrix [2] https://phabricator.wikimedia.org/F52025988
On 2024-05-14 01:17, Volker E. wrote:
Hi everyone, the Design System Team (DST) is proposing the following changes to MediaWiki browser support [1]:
When will you drop this stupid "m" in the URLs? People now link to these "m" URLs where they should use the correct URLs. It's a big shame. It should be a priority.
Example: https://en.wikipedia.org/wiki/Special:LinkSearch?target=en.m.wikipedia.org
On Monday 13 May 2024, Lars Aronsson lars@aronsson.se wrote:
On 2024-05-14 01:17, Volker E. wrote:
Hi everyone, the Design System Team (DST) is proposing the following changes to MediaWiki browser support [1]:
When will you drop this stupid "m" in the URLs? People now link to these "m" URLs where they should use the correct URLs. It's a big shame. It should be a priority.
Example: https://en.wikipedia.org/wiki/Special:LinkSearch?target=en.m.wikipedia.org
-- Lars Aronsson (lars@aronsson.se) Linköping
Can we not hijack threads to complain about totally unrelated things.
-- Bawolff
On Tue, 14 May 2024 at 16:54, Lars Aronsson lars@aronsson.se wrote:
When will you drop this stupid "m" in the URLs? People now link to these "m" URLs where they should use the correct URLs. It's a big shame. It should be a priority.
Example: https://en.wikipedia.org/wiki/Special:LinkSearch?target=en.m.wikipedia.org
-- Lars Aronsson (lars@aronsson.se) Linköping
That would be https://phabricator.wikimedia.org/T214998, which is out of the scope of this email thread, Please keep your tone in consideration when posting to shared lists such as this one.
I very strongly object to dropping Opera. Opera's market share is a lot more than 0.1% on other stats services, there must be something wrong with the detection code. Opera is very popular in Africa and Eastern Europe, and has a lot of users with the Opera GX gaming browser as well. There are also multiple Opera versions as well on a large range of mobile devices. Opera Mini should still be supported as it is useful in areas of the world where data is expensive or slow. I hope you do some more studies and consider keeping Opera support.
Morning Design Team,
I've had a quick look at the usage stats from Miraheze for the last 21 days and I'm seeing around 5.5% of traffic using Opera.
Opera isn't a well used but still fairly popular browser and I'm not sure I believe it's got less 0.1% of traffic for Wikimedia. Where has this data come from?
I have no objection to dropping IE but I do note that it's still about 0.5% for us today. I haven't had time to look at Edge legacy data.
Thanks, Sam
On Tue, 14 May 2024 at 10:44, normanrogers@gmail.com wrote:
I very strongly object to dropping Opera. Opera's market share is a lot more than 0.1% on other stats services, there must be something wrong with the detection code. Opera is very popular in Africa and Eastern Europe, and has a lot of users with the Opera GX gaming browser as well. There are also multiple Opera versions as well on a large range of mobile devices. Opera Mini should still be supported as it is useful in areas of the world where data is expensive or slow. I hope you do some more studies and consider keeping Opera support. _______________________________________________ Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
one thing I'm a little unclear on: does this just refer to non-Chromium Opera, or to Opera as a whole (including the Chromium versions)? if the latter, I'd reconsider - as much as I have my qualms with them, the Chromium Opera(s) have a decently sized user base.
no objections to - if anything, active support for - IE/Edge Legacy being dropped.
kind regards, Rexo (she/they)
On Tue, 14 May 2024 at 12:12, RhinosF1 - rhinosf1@gmail.com wrote:
Morning Design Team,
I've had a quick look at the usage stats from Miraheze for the last 21 days and I'm seeing around 5.5% of traffic using Opera.
Opera isn't a well used but still fairly popular browser and I'm not sure I believe it's got less 0.1% of traffic for Wikimedia. Where has this data come from?
I have no objection to dropping IE but I do note that it's still about 0.5% for us today. I haven't had time to look at Edge legacy data.
Thanks, Sam
On Tue, 14 May 2024 at 10:44, normanrogers@gmail.com wrote:
I very strongly object to dropping Opera. Opera's market share is a lot more than 0.1% on other stats services, there must be something wrong with the detection code. Opera is very popular in Africa and Eastern Europe, and has a lot of users with the Opera GX gaming browser as well. There are also multiple Opera versions as well on a large range of mobile devices. Opera Mini should still be supported as it is useful in areas of the world where data is expensive or slow. I hope you do some more studies and consider keeping Opera support. _______________________________________________ Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org
https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
I suggest we fix https://phabricator.wikimedia.org/T342267 before making any decision that is based on browser usage metric, unless we can demonstrate by other means that Opera isn't used by up to 10% of page views.
By "fix" I mean, ask your respective managers to demand it and create interest in it. I have regularly raised it internally since 2018, but investigating an apparent defect whereby over 10% of global data may be missing, has thus far not been prioritised.
I have no opinion on IE11/Android as those are obviously low on usage without requiring evidence. Given that Opera is Chromium-based, and evergreen, and already listed as "Last N years", I would be skeptical of dropping that solely based on usage data.
Is there a similar gain in CSS/HTML baseline by dropping such entry?
I note that HTML summary/details do not require native support in Basic to adopt. They were specced by WHATWG specifically with progressive enhancement in mind. Browsers render content in unknown elements by default (at least, since IE6).
-- Timo
On Tue, 14 May 2024, at 00:17, Volker E. wrote:
Hi everyone, the Design System Team (DST) is proposing the following changes to MediaWiki browser support [1]:
- Drop support for Internet Explorer 11 (IE 11)
- Drop support for all versions of Edge Legacy
- Drop support for Opera
- Increase Basic (Grade C) support for Chrome and Firefox to versions 49+, Safari and iOS to versions 10+.
What this means: The browsers we’re phasing out won’t be tested for layout rendering anymore. While users on these browsers might and will still be able to read and basically interact with content, they might experience some quirks. This step helps us integrate modern web features more seamlessly.
These changes will unlock the ability to use specific newer browser features that cannot be safely used today without requiring a fallback, notably CSS custom properties (used in upcoming reading customization features like Night Mode) and the <summary> and <details> HTML elements that can be used to replace the checkbox hack. This will reduce the amount of code sent to 99.9% of users and cut down on software development costs and maintenance burdens.
See the full announcement for more details; PDF to download [2].
On behalf of the Wikimedia Foundation Design System Team, Volker [1] https://www.mediawiki.org/wiki/Compatibility#Browser_support_matrix [2] https://phabricator.wikimedia.org/F52025988 _______________________________________________ Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
On Tue, 14 May 2024, at 18:42, Krinkle wrote:
I suggest we fix https://phabricator.wikimedia.org/T342267 before making any decision that is based on browser usage […].
By "fix" I mean, ask your respective managers to [create] demand for it and create interest in it. [it] has thus far not been prioritised. […]
Correction: The task is part of a sprint as of two weeks ago. I failed to see this earlier. Yay!
-- Timo
(it's ok, that sprint is just starting May 23rd, but yes, equally excited to get to work on it)
On Tue, May 21, 2024 at 3:48 PM Krinkle krinkle@fastmail.com wrote:
On Tue, 14 May 2024, at 18:42, Krinkle wrote:
I suggest we fix https://phabricator.wikimedia.org/T342267 before making any decision that is based on browser usage […].
By "fix" I mean, ask your respective managers to [create] demand for it and create interest in it. [it] has thus far not been prioritised. […]
Correction: The task is part of a sprint as of two weeks ago. I failed to see this earlier. Yay!
-- Timo
Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
Update from yesterday: The compatibility template has been updated https://www.mediawiki.org/w/index.php?title=Template%3ACompatibility_browser&diff=6534377&oldid=6503879 to reflect the outlined browser support changes https://www.mediawiki.org/wiki/Compatibility#Browser_support_matrix.
On Tue, May 21, 2024 at 2:32 PM Dan Andreescu dandreescu@wikimedia.org wrote:
(it's ok, that sprint is just starting May 23rd, but yes, equally excited to get to work on it)
On Tue, May 21, 2024 at 3:48 PM Krinkle krinkle@fastmail.com wrote:
On Tue, 14 May 2024, at 18:42, Krinkle wrote:
I suggest we fix https://phabricator.wikimedia.org/T342267 before making any decision that is based on browser usage […].
By "fix" I mean, ask your respective managers to [create] demand for it and create interest in it. [it] has thus far not been prioritised. […]
Correction: The task is part of a sprint as of two weeks ago. I failed to see this earlier. Yay!
-- Timo
Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org
https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
I'd like to add my belated support for removing Opera from the compatibility table. Today's Opera is just another Chromium reskin, and it doesn't make sense to list it separately, just like we don't list Brave or Vivaldi or any of the other less-known Chromium reskins.
(And as a former Opera fan, I'd like to add: if you're still using it out of nostalgia, it's high time to switch. The company currently behind it doesn't inspire trust in their engineering. I recommend Firefox.)
Keeping the archives happy:
On Tue, May 14, 2024 at 1:43 PM Krinkle krinkle@fastmail.com wrote:
I suggest we fix https://phabricator.wikimedia.org/T342267 before making any decision that is based on browser usage metric
This has happened, new numbers look better https://browser-reports.wmcloud.org/ for purposes such as you all discuss here.
I'm assuming any change to drop Opera off basic and modern support relates to the legacy Opera, not Chromium based Operas, just like how the Chromium Edge versions still get modern JavaScript and CSS?
Probably worth calling this out explicitly.
-- brooke
On Mon, May 13, 2024, 4:18 PM Volker E. volker.e@wikimedia.org wrote:
Hi everyone, the Design System Team (DST) is proposing the following changes to MediaWiki browser support [1]:
- Drop support for Internet Explorer 11 (IE 11)
- Drop support for all versions of Edge Legacy
- Drop support for Opera
- Increase Basic (Grade C) support for Chrome and Firefox to versions 49+,
Safari and iOS to versions 10+.
What this means: The browsers we’re phasing out won’t be tested for layout rendering anymore. While users on these browsers might and will still be able to read and basically interact with content, they might experience some quirks. This step helps us integrate modern web features more seamlessly.
These changes will unlock the ability to use specific newer browser features that cannot be safely used today without requiring a fallback, notably CSS custom properties (used in upcoming reading customization features like Night Mode) and the <summary> and <details> HTML elements that can be used to replace the checkbox hack. This will reduce the amount of code sent to 99.9% of users and cut down on software development costs and maintenance burdens.
See the full announcement for more details; PDF to download [2].
On behalf of the Wikimedia Foundation Design System Team, Volker
[1] https://www.mediawiki.org/wiki/Compatibility#Browser_support_matrix [2] https://phabricator.wikimedia.org/F52025988 _______________________________________________ Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
Hi all, agreed Brooke and all. There was one formulation in my starting email that should have phrased slightly differently for more clarity.
'Drop support' should have been 'drop dedicated support'. where users of Opera continue to be able to browse or edit the wikis in a basic manner. These changes, starting with the support matrix, just mean we're not actively testing against them anymore or, specifically for IE11, going a long way to ensure complete layout rendering equivalence with modern browsers. Again, in the case of Opera, Chromium versions will probably see exactly the same experience as before through active testing and bug fixing for supported versions of Chrome.
Per Timo's note about https://phabricator.wikimedia.org/T342267 and in response to Sam's question, some data sets like those used by public analytics tools like https://analytics.wikimedia.org/dashboards/browsers/#all-sites-by-browser are not believed to be completely accurate at this time. We used only Foundation internally accessible https://superset.wikimedia.org/superset/explore/p/lj9Q8J5b67K/, which to our knowledge doesn't suffer from the same concerns about accuracy. This data suggests that over the past 12 months, usage across all Opera browsers combined comes to just over 0.1%, significantly less than even Edge Legacy. The updated matrix would reflect the reality that Opera is not actively tested against when developing new features.
Best, Volker
On Tue, May 14, 2024 at 2:48 PM Brooke Vibber bvibber@wikimedia.org wrote:
I'm assuming any change to drop Opera off basic and modern support relates to the legacy Opera, not Chromium based Operas, just like how the Chromium Edge versions still get modern JavaScript and CSS?
Probably worth calling this out explicitly.
-- brooke
On Mon, May 13, 2024, 4:18 PM Volker E. volker.e@wikimedia.org wrote:
Hi everyone, the Design System Team (DST) is proposing the following changes to MediaWiki browser support [1]:
- Drop support for Internet Explorer 11 (IE 11)
- Drop support for all versions of Edge Legacy
- Drop support for Opera
- Increase Basic (Grade C) support for Chrome and Firefox to versions
49+, Safari and iOS to versions 10+.
What this means: The browsers we’re phasing out won’t be tested for layout rendering anymore. While users on these browsers might and will still be able to read and basically interact with content, they might experience some quirks. This step helps us integrate modern web features more seamlessly.
These changes will unlock the ability to use specific newer browser features that cannot be safely used today without requiring a fallback, notably CSS custom properties (used in upcoming reading customization features like Night Mode) and the <summary> and <details> HTML elements that can be used to replace the checkbox hack. This will reduce the amount of code sent to 99.9% of users and cut down on software development costs and maintenance burdens.
See the full announcement for more details; PDF to download [2].
On behalf of the Wikimedia Foundation Design System Team, Volker
[1] https://www.mediawiki.org/wiki/Compatibility#Browser_support_matrix [2] https://phabricator.wikimedia.org/F52025988 _______________________________________________ Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org
https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
wikitech-l@lists.wikimedia.org