powidl.eu doesn't render correctly with Firefox on Linux, because of font fallback
Categories
(Core :: Layout, defect)
Tracking
()
People
(Reporter: markus.popp, Unassigned)
Details
Attachments
(2 files)
Open http://www.powidl.eu in Firefox and a Chromium based browser and compare the layout.
The layout is broken in Firefox.
Reporter | ||
Comment 1•5 years ago
|
||
Comment 2•5 years ago
|
||
I think this is WONTFIX. This works on Windows because the page specifies font-family: Tahoma
and relies on the font chosen to have the same line breaking.
Since Tahoma is not installed on Linux, they effectively just use the default font, which ends up being larger on Linux.
Firefox uses DejaVu Serif
on my system as a fallback, Chrome seems to be using Nimbus Roman
. fc-match serif
returns DejaVu on my system so Firefox seems a bit closer to system behavior here...
Other than changing default serif font on Linux to be Nimbus Roman
rather than DejaVu it's unclear what we can do here.
Jonathan any idea how Chrome may come up with Nimbus for this case?
Comment 3•5 years ago
|
||
Can you confirm that changing font settings from about:preferences
to make the serif
family something smaller like Nimbus Roman
fixes the website on your end too?
Updated•5 years ago
|
Reporter | ||
Comment 4•5 years ago
|
||
(In reply to Emilio Cobos Álvarez (:emilio) from comment #3)
Can you confirm that changing font settings from
about:preferences
to make theserif
family something smaller likeNimbus Roman
fixes the website on your end too?
This works. Now it looks more like it does in Chrome.
Comment 5•5 years ago
|
||
(In reply to Emilio Cobos Álvarez (:emilio) from comment #2)
I think this is WONTFIX. This works on Windows because the page specifies
font-family: Tahoma
and relies on the font chosen to have the same line breaking.Since Tahoma is not installed on Linux, they effectively just use the default font, which ends up being larger on Linux.
Firefox uses
DejaVu Serif
on my system as a fallback, Chrome seems to be usingNimbus Roman
.fc-match serif
returns DejaVu on my system so Firefox seems a bit closer to system behavior here...Other than changing default serif font on Linux to be
Nimbus Roman
rather than DejaVu it's unclear what we can do here.Jonathan any idea how Chrome may come up with Nimbus for this case?
I don't know specifically, but I'd assume this is coming from their equivalent of our default font prefs or something like that.
I notice that they seem to have a hard-coded default of "Times New Roman" in https://chromium.googlesource.com/chromium/src/+/refs/heads/master/content/public/common/web_preferences.cc#234, and it wouldn't surprise me if fontconfig is configured to substitute family-name "Times New Roman" with Nimbus if it's installed. (What does fc-match "Times New Roman"
return for you?)
Comment 6•5 years ago
|
||
So this site's layout is terribly inflexible, and makes unreliable assumptions about the user's browser environment. It has similar issues even on a Windows or Mac system with Tahoma available if I set a minimum font size such as 16px in the browser prefs (as a user with limited vision might do, for example). And the same thing applies in Chrome if I set a min font size in its preferences, too.
I'd consider this INVALID as a browser bug report, but it could be reported to the site as a compatibility and accessibility issue with their design.
Comment 7•5 years ago
|
||
(In reply to Jonathan Kew (:jfkthame) from comment #5)
I notice that they seem to have a hard-coded default of "Times New Roman" in https://chromium.googlesource.com/chromium/src/+/refs/heads/master/content/public/common/web_preferences.cc#234, and it wouldn't surprise me if fontconfig is configured to substitute family-name "Times New Roman" with Nimbus if it's installed. (What does
fc-match "Times New Roman"
return for you?)
Ah, I failed to spot that line. Yeah that sounds about right.
Comment 8•5 years ago
|
||
I'd consider this INVALID as a browser bug report, but it could be reported to the site as a compatibility and accessibility issue with their design.
Yeah, I agree given comment 4.
Description
•