Open Bug 1647833 Opened 4 years ago Updated 4 years ago

Improve address bar focus state accessibility for light sensitivity

Categories

(Firefox :: Address Bar, enhancement, P3)

enhancement

Tracking

()

People

(Reporter: ntim, Unassigned)

References

(Blocks 1 open bug)

Details

(5 keywords)

Attachments

(2 files)

Attached image Firefox screenshot (deleted) —

Bug 1629303 only removed the animation on focus, but the animation is unrelated to the flickering light problem IMO. The flickering light is due to the popup being lighter, using bright colors and popping up on focus by surprise:

  • The popup is brighter in than the rest of the UI to catch attention
  • The Switch to Tab color is fairly bright
  • The highlight color is shown on the selected result and the border (it probably doesn't help that I'm using a bright pink, but still, Chrome uses a shadow and displays a different shade of gray for selected results)
  • It doesn't help that the popup expands and appears suddenly directly on focus, to catch attention, so it makes it easy to accidentally trigger the popup. You could argue that users can disable top sites, but it would be a shame if users can't use a feature for accessibility reasons.
  • FWIW, icons tend to be bright as well, so it doesn't help to have 10 showing up on focus

None of these issues are present on other browsers.

Attached image Chrome screenshot (deleted) —

flickering light problem was not the focus of my bug. My bug was the expanding/shrinking and how it relates to epilepsy. With the fix to adhere to the reduced animation OS has help a great deal. I don't see the flashing/flicker problem as I don't use custom colors/themes.

I think it's worth to also point out Edge solution to show a grey background with a selection color border-left.

(In reply to Troy Janda from comment #2)

flickering light problem was not the focus of my bug. My bug was the expanding/shrinking and how it relates to epilepsy. With the fix to adhere to the reduced animation OS has help a great deal. I don't see the flashing/flicker problem as I don't use custom colors/themes.

Custom themes are kind of moot since a person with some kind of color sensibility wouldn't choose a theme with pink as the highlight color. I'll mark this S4 as it doesn't seem we've established that there's an outstanding accessibility problem, let alone how severe it might be.

No longer blocks: 1629303
Severity: -- → S4
Type: task → enhancement
Depends on: 1629303
Keywords: access, blocked-ux
Priority: -- → P5

(In reply to Dão Gottwald [::dao] from comment #4)

(In reply to Troy Janda from comment #2)

flickering light problem was not the focus of my bug. My bug was the expanding/shrinking and how it relates to epilepsy. With the fix to adhere to the reduced animation OS has help a great deal. I don't see the flashing/flicker problem as I don't use custom colors/themes.

Custom themes are kind of moot since a person with some kind of color sensibility wouldn't choose a theme with pink as the highlight color. I'll mark this S4 as it doesn't seem we've established that there's an outstanding accessibility problem, let alone how severe it might be.

The default highlight color is blue, and is even the graphite (gray) color is fairly light.

Again, regardless the highlight color choice, Chrome/Edge/Safari are all doing a better job here.

(In reply to Tim Nguyen :ntim (please use needinfo?) from comment #5)

(In reply to Dão Gottwald [::dao] from comment #4)

(In reply to Troy Janda from comment #2)

flickering light problem was not the focus of my bug. My bug was the expanding/shrinking and how it relates to epilepsy. With the fix to adhere to the reduced animation OS has help a great deal. I don't see the flashing/flicker problem as I don't use custom colors/themes.

Custom themes are kind of moot since a person with some kind of color sensibility wouldn't choose a theme with pink as the highlight color. I'll mark this S4 as it doesn't seem we've established that there's an outstanding accessibility problem, let alone how severe it might be.

The default highlight color is blue, and is even the graphite (gray) color is fairly light.

If we're talking about people with a clinical condition, we can expect those people to change settings to better cater to their needs. Which is to say, I'm not sure that the default matters much in this context. The graphite case seems more interesting, but ultimately those people might be served best by using an OS that allows for more customization...

Again, regardless the highlight color choice, Chrome/Edge/Safari are all doing a better job here.

If you're starting with the premise that light colors are problematic per se, then yes, they may be doing a better job. However, I still don't think we have established that there's an accessibility problem in the first place. In fact, for people with poor vision the higher contrast is likely beneficial. (Except for the fact that our popup is lighter in the dark theme, which is indeed counterproductive.)

(In reply to Dão Gottwald [::dao] from comment #6)

If we're talking about people with a clinical condition, we can expect those people to change settings to better cater to their needs. Which is to say, I'm not sure that the default matters much in this context. The graphite case seems more interesting, but ultimately those people might be served best by using an OS that allows for more customization...

I think your expectation that people with an diagnosed, or even un-diagnosed (remember, in the US because of horrible healthcare, people go without diagnosis) condition of motion/light senstitivity would know to enable or choose OS settings to mitigate harm is a big assumption.

Secondly, many people will not have the option to choose an OS because they can't afford to, don't have the resources (time, cognitive surpluses) to learn a new OS, or may have their choice of OS and even settings determined by an employer or institution.

(In reply to Emma Humphries, Bugmaster ☕️🎸🧞‍♀️✨ (she/they) [:emceeaich] (UTC-8) needinfo? me from comment #7)

I think your expectation that people with an diagnosed, or even un-diagnosed (remember, in the US because of horrible healthcare, people go without diagnosis) condition of motion/light senstitivity would know to enable or choose OS settings to mitigate harm is a big assumption.

Secondly, many people will not have the option to choose an OS because they can't afford to, don't have the resources (time, cognitive surpluses) to learn a new OS, or may have their choice of OS and even settings determined by an employer or institution.

I didn't mean to assert that all potentially affected users will choose custom colors or different OS, but in the end this is really the only way for all users to be catered for. When some users need to avoid light colors and others need increased contrast, it's impossible to optimize for all at the same time. Maybe light sensitivity is more prevalent and should be catered for first, but I don't know that; we only seem to be talking hypothetically so far with no evidence that there's an accessibility problem at all.

(In reply to Dão Gottwald [::dao] from comment #8)

I didn't mean to assert that all potentially affected users will choose custom colors or different OS, but in the end this is really the only way for all users to be catered for. When some users need to avoid light colors and others need increased contrast, it's impossible to optimize for all at the same time. Maybe light sensitivity is more prevalent and should be catered for first, but I don't know that; we only seem to be talking hypothetically so far with no evidence that there's an accessibility problem at all.

All EEGs (part of epilepsy diagnosis) have light sensitivity tests, which are flickering lights starting at very low frequency and accelerating until a seizure happens, so yes, light sensitivity is an accessibility problem, see this page on photosensitive epilepsy. I've also learnt that there's other conditions related to light sensitivity but unrelated to epilepsy.

Now, adjustments might be made in the lives of photosensitive folks, but that doesn't justify putting popups (especially with lots of colors/light) that can be triggered too easily, especially by default. As Emma mentioned, not everyone knows how to do those adjustments themselves either, and may find it easier to switch to a different browser.

I did not question whether light sensitivity is real but 1) how prevalent it is and whether it should by default be optimized for with priority over bad vision (I doubt that; according to the page you linked to, about 0.03% of the population are affected), and 2) whether the points you listed in comment 0 are in fact triggers (I doubt that too).

(In reply to Dão Gottwald [::dao] from comment #10)

I did not question whether light sensitivity is real but 1) how prevalent it is and whether it should by default be optimized for with priority over bad vision (I doubt that; according to the page you linked to, about 0.03% of the population are affected),

I'm not asking one to be prioritized over the other. It is possible to design for both, through simple changes (like other browsers), and through the prefers-contrast media query which is coming (bug 1506364).

and 2) whether the points you listed in comment 0 are in fact triggers (I doubt that too).

Might be an exaggerated image, but think of the popup as a flashlight, individually: being bright, having multiple colors, showing up suddenly, it may not be the worse problem having it turn on in front of you, but accumulated, it is uncomfortable (even for a non-epileptic person tbh).

The only solution to the concerns you pointed at seems to be a complete recoloring, that is something UX may evaluate. Though, finding a proper solution requires an actual study of the problem. For example we don't know if the background color is a problem, if the action text color is a problem, or if it's the selection. Or maybe the problem is minimal because the flicker frecency is too low.
Short, we just don't know and are mostly making assumptions here. To solve that, we'd need a user study done on light sensitive users. Accessibility expressed some interest in those kind of studies, but it's not trivial to setup them.

(In reply to Tim Nguyen :ntim (unavailable) from comment #11)

(In reply to Dão Gottwald [::dao] from comment #10)

I did not question whether light sensitivity is real but 1) how prevalent it is and whether it should by default be optimized for with priority over bad vision (I doubt that; according to the page you linked to, about 0.03% of the population are affected),

I'm not asking one to be prioritized over the other. It is possible to design for both, through simple changes (like other browsers),

We have specific product and UX goals that are often different from other browsers. E.g. we want to direct users to the address bar whereas Chrome wants to drive users to google.com.

You also mentioned the Switch to Tab label color, which we explicitly made lighter to improve contrast. So this is directly at odds with what you want.

and through the prefers-contrast media query which is coming (bug 1506364).

I expect low vision where contrast helps to be much more prevalent, and as we all know, not everyone affected will opt into high contrast modes. So we can use prefers-contrast to enhance contrast further, but by default we still need to adhere to standard contrast requirements.

I think the most obvious and (if we can convince verdi) least controversial improvement we could make here is to darken the popup background, which would immediately allow us to tune down the action label colors as well.

Blocks: 1630275
Summary: Improve accessibility of address bar focus state → Improve address bar focus state accessibility for light sensitivity
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: