rAc: remove bold section labels on linux_aura.

rAc: remove bold section labels on linux_aura.

R=estade@chromium.org
BUG=293913
TEST=visual

[Dan Beam, 2014-01-22 01:40:42]

(screenshot here:
https://code.google.com/p/chromium/issues/detail?id=293913#c29)

[Evan Stade, 2014-01-22 01:56:06]

what happened to my suggestion from the bug? a ui/ function for
"BoldOnlyIfItLooksGood"?

[Dan Beam, 2014-01-22 03:01:53]

On 2014/01/22 01:56:06, Evan Stade wrote:
> what happened to my suggestion from the bug? a ui/ function for
> "BoldOnlyIfItLooksGood"?

do you have other clients of this API in mind?  I figured "IfItLooksGood" is up
to each project.

[Evan Stade, 2014-01-22 20:10:56]

On 2014/01/22 03:01:53, Dan Beam wrote:
> On 2014/01/22 01:56:06, Evan Stade wrote:
> > what happened to my suggestion from the bug? a ui/ function for
> > "BoldOnlyIfItLooksGood"?
> 
> do you have other clients of this API in mind?  I figured "IfItLooksGood" is
up
> to each project.

on the contrary, I think that would lead to a great deal of inconsistency if
each project had to independently realize this is a good way to do things and
assess each platform for visual appeal

[Dan Beam, 2014-01-23 01:27:57]

On 2014/01/22 20:10:56, Evan Stade wrote:
> On 2014/01/22 03:01:53, Dan Beam wrote:
> > On 2014/01/22 01:56:06, Evan Stade wrote:
> > > what happened to my suggestion from the bug? a ui/ function for
> > > "BoldOnlyIfItLooksGood"?
> > 
> > do you have other clients of this API in mind?  I figured "IfItLooksGood" is
> up
> > to each project.
> 
> on the contrary, I think that would lead to a great deal of inconsistency if
> each project had to independently realize this is a good way to do things and
> assess each platform for visual appeal

Done.

[Dan Beam, 2014-01-23 03:11:24]

+thakis@ for chrome/browser/ui/cocoa

[Nico, 2014-01-23 03:13:55]

cocoa lgtm

api lgmeh

https://chromiumcodereview.appspot.com/144073004/diff/160001/chrome/browser/u...
File chrome/browser/ui/cocoa/autofill/autofill_section_container.mm (right):

https://chromiumcodereview.appspot.com/144073004/diff/160001/chrome/browser/u...
chrome/browser/ui/cocoa/autofill/autofill_section_container.mm:481: if
(label.font & ui::ResourceBundle::BoldFont) {
bitmasks in an api are bad api

[Dan Beam, 2014-01-23 03:17:45]

+tony@ for ui/base/resource

[tony, 2014-01-23 17:38:25]

LGTM

FWIW, I liked dbeam's original code better, although I'm not sure what
specifically about the Linux Aura bold doesn't "look good".  Is it because it's
fuzzy?

Also, does the bold style on Linux Aura match the rest of system?  That is,
maybe that's what bold is supposed to look like and we should follow the
platform.  If not, maybe it's a bug in how we're computing bold.  Does Linux
Aura obey the user's antialiasing and hinting settings?

[Dan Beam, 2014-01-23 17:48:30]

On 2014/01/23 17:38:25, tony wrote:
> LGTM
> 
> FWIW, I liked dbeam's original code better, although I'm not sure what
> specifically about the Linux Aura bold doesn't "look good".  Is it because
it's
> fuzzy?
> 
> Also, does the bold style on Linux Aura match the rest of system?  That is,
> maybe that's what bold is supposed to look like and we should follow the
> platform.  If not, maybe it's a bug in how we're computing bold.  Does Linux
> Aura obey the user's antialiasing and hinting settings?

maybe the screenshots [1] can answer these questions.  i think it's more an
issue of the kerning of the linux aura default font.

[1] https://code.google.com/p/chromium/issues/detail?id=293913#c25

[tony, 2014-01-23 17:56:21]

On 2014/01/23 17:48:30, Dan Beam wrote:
> On 2014/01/23 17:38:25, tony wrote:
> > LGTM
> > 
> > FWIW, I liked dbeam's original code better, although I'm not sure what
> > specifically about the Linux Aura bold doesn't "look good".  Is it because
> it's
> > fuzzy?
> > 
> > Also, does the bold style on Linux Aura match the rest of system?  That is,
> > maybe that's what bold is supposed to look like and we should follow the
> > platform.  If not, maybe it's a bug in how we're computing bold.  Does Linux
> > Aura obey the user's antialiasing and hinting settings?
> 
> maybe the screenshots [1] can answer these questions.  i think it's more an
> issue of the kerning of the linux aura default font.
> 
> [1] https://code.google.com/p/chromium/issues/detail?id=293913#c25

I looked at the screen shots.  Is it that the letters are too close together or
are there specific pairs of letters that you think stand out?  If the user
changes their system font, does it look better?  Is it any different from bold
in HTML?

I think the kerning isn't perfect, but seems about the same as the the CrOS one.
 I'd be surprised if the Linux version and the CrOS version would be different
if both were at the same DPI (e.g., CrOS on any device other than a Pixel). 
Aren't they using the same code path?

[Evan Stade, 2014-01-23 19:30:34]

On 2014/01/23 17:56:21, tony wrote:
> On 2014/01/23 17:48:30, Dan Beam wrote:
> > On 2014/01/23 17:38:25, tony wrote:
> > > LGTM
> > > 
> > > FWIW, I liked dbeam's original code better, although I'm not sure what
> > > specifically about the Linux Aura bold doesn't "look good".  Is it because
> > it's
> > > fuzzy?
> > > 
> > > Also, does the bold style on Linux Aura match the rest of system?  That
is,
> > > maybe that's what bold is supposed to look like and we should follow the
> > > platform.  If not, maybe it's a bug in how we're computing bold.  Does
Linux
> > > Aura obey the user's antialiasing and hinting settings?
> > 
> > maybe the screenshots [1] can answer these questions.  i think it's more an
> > issue of the kerning of the linux aura default font.
> > 
> > [1] https://code.google.com/p/chromium/issues/detail?id=293913#c25
> 
> I looked at the screen shots.  Is it that the letters are too close together
or
> are there specific pairs of letters that you think stand out?  If the user
> changes their system font, does it look better?  Is it any different from bold
> in HTML?

The decision making process occurred on the linked bug. Questions like these
should be directed to Jeff Stark.

> 
> I think the kerning isn't perfect, but seems about the same as the the CrOS
one.
>  I'd be surprised if the Linux version and the CrOS version would be different
> if both were at the same DPI (e.g., CrOS on any device other than a Pixel). 
> Aren't they using the same code path?

fair enough, Dan can get a better screenshot for CrOS?

[Dan Beam, 2014-01-24 21:57:32]

On 2014/01/23 19:30:34, Evan Stade wrote:
> On 2014/01/23 17:56:21, tony wrote:
> > On 2014/01/23 17:48:30, Dan Beam wrote:
> > > On 2014/01/23 17:38:25, tony wrote:
> > > > LGTM
> > > > 
> > > > FWIW, I liked dbeam's original code better, although I'm not sure what
> > > > specifically about the Linux Aura bold doesn't "look good".  Is it
because
> > > it's
> > > > fuzzy?
> > > > 
> > > > Also, does the bold style on Linux Aura match the rest of system?  That
> is,
> > > > maybe that's what bold is supposed to look like and we should follow the
> > > > platform.  If not, maybe it's a bug in how we're computing bold.  Does
> Linux
> > > > Aura obey the user's antialiasing and hinting settings?
> > > 
> > > maybe the screenshots [1] can answer these questions.  i think it's more
an
> > > issue of the kerning of the linux aura default font.
> > > 
> > > [1] https://code.google.com/p/chromium/issues/detail?id=293913#c25
> > 
> > I looked at the screen shots.  Is it that the letters are too close together
> or
> > are there specific pairs of letters that you think stand out?  If the user
> > changes their system font, does it look better?  Is it any different from
bold
> > in HTML?
> 
> The decision making process occurred on the linked bug. Questions like these
> should be directed to Jeff Stark.
> 
> > 
> > I think the kerning isn't perfect, but seems about the same as the the CrOS
> one.
> >  I'd be surprised if the Linux version and the CrOS version would be
different
> > if both were at the same DPI (e.g., CrOS on any device other than a Pixel). 
> > Aren't they using the same code path?
> 
> fair enough, Dan can get a better screenshot for CrOS?

https://code.google.com/p/chromium/issues/detail?id=293913#c32

[Dan Beam, 2014-01-27 22:01:26]

ping estade@