Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
macOS Tahoe: A Pattern of Poor Usability and Design Regression
- Thread starter Deuce2
- Start date
- Sort by reaction score
It is implicit that everyone expresses their opinion. This insistence on clarifications wastes space on the server and nothing else. Facts, concreteness, that's all that's asked for... I like the new spotlight because it allows me to do... I love the different angles because they improve my workflow because...
Kal Madda
macrumors 601
Those are user customizable. The person customizing that contact card chose to use white text on a light peach colored background. The system shouldn’t undo user customizations, or then it wouldn’t be truly customizable. What would the point of having text and background customization options for contact cards if the system just changed the text or background color you chose? I don’t get how or why Apple is supposed to change user customizations for their contact cards. The status bar is one thing, as that is a system element. But someone custom contact card won’t be so custom if the system enforces color changes onto it… 🤷🏼♂️🤔
How do you know the highlighted claim to be true? You have rigorous statistical evidence for this claim? Like a full scientific survey with a sufficient sample of all Apple users and their thoughts about the legibility? Or are you just making assumptions based on your limited experience? 🤔
And it is an opinion. 🤷🏼♂️
Kal Madda
macrumors 601
Yet many try to claim or represent their opinions as “facts”… 🤷🏼♂️. So that can and should be pushed back on…
It seems like a full-time job, and I'm glad it's done with enjoyment and obsession. If the same commitment applied to the second (ignored) part of the post, we would all be happy.
avkills
macrumors 65816
Comparing print to computer monitors is not relevant in this discussion. The way light interacts and our eyes and brain process both are completely different; hence the enormous attempt to make computer displays match print for designers. CMYK is just different than RGBA.
Trying to apply printed design philosophy to computer graphic design almost always turns out bad.
Kal Madda
macrumors 601
People view book covers digitally as well… 🧐. Many designs are viewed both ways these days… 🤷🏼♂️
fisherking
macrumors G5
if only it were that simple. the problem is people who say things like: "everyone agrees", "it's a fact that"... people who speak in absolutes, or think that they and their 3 friends represent everyone 😳
avkills
macrumors 65816
All I am getting at is that print is usually printed on white; computer graphics starts on black. My experience is with large scale graphics being projected or thrown to LED walls. Usually what works for print does not work well in this scenario. Interesting thing about screens, is that is also white to start unless rear project, and then that becomes the black. LED walls are better; but heavier and power hungry.
It is all about making it not hurt your eyes to look at it; one of the reasons why I really hate Apple's policy on high refresh displays. This automatic **** is stupid. Just let us run the display at higher than 60hz all the time.
Sometimes you need the display at 60hz (or 59.94, which Apple also hates) like if you are monitoring video or something, but high refresh on the main display with the GUI just makes it easier on the eyes.
Just my take on it.
star-affinity
macrumors 68020
OK, that is more than I know…
I can't seem to change the font color for a contact card – there are only backgrounds and fonts and font sizes.
Of course I don't have statistical evidence ”on paper”. But I see the talks about it online and at the digital design studio where I work. So while not the evidence you ask for it surely has some real substance. But it is my impression, not a proven fact.
Still, that legibility and contrast is lower than in previous macOS versions and that those factors are dependent on the background to this extent is new to Apple operating systems. The two screenshots from @Tig_one on page three of this thread clearly shows how parts of the GUI elements merges quite a lot with the background and prone to more confusion than before.
That you think this looks fine doesn't mean that legibility and contrast isn't objectively less than in previews Apple operating systems.
So yes, opinions differ on whether it’s a problem in practice. But the underlying observation – that contrast is lower in some scenarios – isn’t an opinion. That’s the distinction I’m trying to make.
avkills
macrumors 65816
The last image in this post is a design fail in my opinion. Content that is interactive should not be shown under OS buttons that are also interactive.
Kal Madda
macrumors 601
In the Contacts app, you can go to your contact card, and right at the top of the contact card should be your avatar. Below that is the “edit” button. If you tap on that, you get various customization options. Select the “Poster” option at the top. And in the poster, if you tap the text, it gives you customization options for the fonts, size, and a color option. I don’t know if you can change other people’s contact cards if they customized it poorly, but you can definitely customize your own. And contact cards don’t have poor contrast by default, people have to customize them to make them that way.
Okay, well at least we can separate that out as an impression. I don’t share that impression. 🙂👍🏻
Not really as much as you seem to think. Transparency isn’t completely new to the Mac. And contrast of Liquid Glass is typically higher than examples like this ‘Notifications” button from macOS Catalina:
It isn’t “objectively less”, you just think it is… 🤷🏼♂️
That is an opinion… Because I could find several examples of “low contrast” elements like the one included above in previous versions of macOS as well…
Kal Madda
macrumors 601
So I did an interesting test with the same wallpaper from Catalina on macOS Tahoe, and the Liquid Glass Control Center in the same upper right position alongside a screenshot of Notification Center in macOS Catalina. I compared the contrast for the Liquid Glass buttons vs the contrast of the “Notifications” button on macOS Catalina. The lightest background color that bled through with Liquid Glass still provided nearly double the contrast with Liquid Glass according to Figma’s color contrast accessibility tool. Even Liquid Glass buttons over bright white backdrops had higher contrast. And these are stock settings. People can increase contrast with accessibility settings if needed. But I found that quite interesting considering all the claims of “poor legibility” with Tahoe in threads like this in this forum, and claims that “Apple’s never used such ‘low contrast’ UI elements”… I think people are just being more nitpicky and specifically looking for things to call “low contrast” in Tahoe, since it’s a new design… 🤷🏼♂️
star-affinity
macrumors 68020
Aha, got it! That was quite a few steps, but now I see where you can change the font color for the contact.
Now I fully understand why this can happen for contacts. Stand corrected on this one – not Apple's fault! 😀
OK. Interesting...
OK, there are bad contrast examples in previous macOS versions (I guess they where heading for Liquid Glass…😀)
I do actually remember thinking notification area looking a bit off compared to the rest of the OS long time ago (not sure when it started).
I shouldn't have said it's entirely new then. But there are more examples in Tahoe. I mean, I've never had an issue with how the App Switcher displays text under each app. In Tahoe it has become more content dependent (like I showed). How the Control Center looks and is more background/content dependent for legibility is new.
I think we’ve reached the point where we’re fundamentally using the word “objective” differently – and that’s okay.
I guess you know what I mean, but when I say contrast is objectively lower in some scenarios, I’m using “objective” in the design/engineering sense: given two foreground/background color pairs, their contrast ratio can be calculated and compared. In some macOS 26 UI states, those ratios are lower than in comparable states in earlier versions. That’s the narrow claim I’m making.
I’m not claiming this is entirely new to macOS, nor that older versions never had low-contrast elements. I’m saying that some newer design choices make contrast more dependent on underlying content, which increases variability in legibility.
You’re absolutely right that whether this matters in practice is subjective, and clearly it doesn’t affect you the same way it affects me. At that point, we’re just describing different experiences – not something one of us can “prove” to the other.
I think we’ve clarified our positions well, so I’m happy to leave it there.
Hope the weekend was good and let's prepare for the onslaught of a new working week! 😀
And maybe macOS 26.3 that fixes all my concerns? 😛 😀
svenmany
macrumors 68030
You have no experience in web development, except maybe you took a class. Without user feedback related to UI choices, directly or indirectly (though a product manager or peers), a class is equivalent to no experience for the topic we're discussing.
You've disagreed twice with me regarding the impact of the corner rounding on web development. It's puzzling that someone with no experience should express any beliefs on these things. The corner issue actually touches on some things a general designer should be sensitive to. You claimed to be a designer, so it's surprising even that eludes you.
Fastmail provides more than a email service that you can access with a general mail program, like Apple Mail. They also offer a web application that is very powerful. I access Fastmail almost exclusively from a web browser. It's a very well-done web application and their handling of corner issues makes it a good case study.
Focusing first on Chrome running on Windows (which would be the largest part of their user base), here's what the bottom right corner looks like:
At this point you wonder what I'm talking about regarding corner issues. Well, that's only the way the corner looks if the width of the region showing mail content is less that 1152 pixels. If you enlarge the window you would see:
The achieved this with a simple media query which you can see in any browser's developer tools:
The Fastmail UI designers paid very careful attention to the edges and corners. They decided that their users would benefit from no border if width were constrained, giving the users more room to work. But if there is room, they add a border. But they didn't do it casually; they adjusted the curvature radius of the border to match Chrome's. They even went as far as to pick a margin of 8px for light mode versus 7px for dark mode. The corner curvature of Chrome on Windows roughly matched the curvature on Sequoia's Safari. They really cared about the corners. All was good until Tahoe.
They had to suffer some loss of that perfection for users running their web app on Tahoe. All browsers seem to have increased their radius of curvature there, but to varying degrees. It's kind of hard to see the defect in Firefox on Tahoe. I can see the defect but you won't be able to. The curvatures don't match. But, I like it still because it creates a constant width border as one rounds the corner. So it's easy to argue it's not really a defect (depending on the designers' goals).
When we get to the latest Chrome, where Google started to cater for Tahoe corner rounding conventions, the defect is obvious:
When they encounter Safari on Tahoe, their goal is completely unachievable. Here's how it looks there:
Consistent border effect cannot be achieved without tuning for each browser and version. I haven't been asked to solve this, so I haven't done any research into it. When I first thought about the problem I wondered about the "sec-ch-ua-*" HTTP headers. These allow you to query detailed browser version information but aren't supported by Safari. (https://developer.mozilla.org/en-US/docs/Web/HTTP/Reference/Headers/Sec-CH-UA). Interested people might want to read more here:
At this point, I hope other readers notice that serious UI designers have to address corner issues and that radical variations in browser window corners introduce challenges. I have a feeling you'll just offer some counter argument. It's a puzzle.
Some time ago, when you were going on about how you were proud to be an American, you said that you were young. That explains some of your tendency to undervalue experience. You also said that you were just a freelance designer. Not having a full-time job with a mentor and regular peers explains some more. You have nothing to draw from except an intuition that has only your own preferences as a foundation. When I mentioned the NIH paper at https://pmc.ncbi.nlm.nih.gov/articles/PMC4975689/, you said
This is a very old topic and wasn't news to me. It wouldn't have been news to you had you been working professionally with other designers. And I'm sure I can find some other reference that discusses the tension introduced when juxtaposing a hard angle up against a rounded curve, as we have with the left bottom corner of the latest Pages. I'm not going to bother looking.
For those without experience, this thread is easily confused with the "Do you like you like Liquid Glass on Mac" thread. You can see that confusion in post #7 of this thread where the poster said "good thing you started another thread about these things". That first thread is oriented toward an individual's opinions and preferences. This thread is about design issues found in Tahoe. Based on your intelligence, I have a feeling you'd have something relevant to offer if you had some experience in the area. As it stands now, this thread has just become a clone of the other thread. It's unfortunate.
Kal Madda
macrumors 601
The usual response to a well-articulated post is generally this:
It's a relevant offer... in its own way.
Edit:
... and the fact that it was modified (not improved much, however) after this post shows at least some shame... perhaps there is hope.
Last edited:
fisherking
macrumors G5
it's true! (i get a good number of those responses to my reasonable posts) 🙄The usual response to a well-articulated post is generally this:
It's a relevant offer... in its own way.
Last edited:
It's a rather worrying trend: instead of discussing things, when there are no arguments, people resort to mockery. I think this should be avoided after the age of 6.
fisherking
macrumors G5
i agree, but it's tricky, as we use that same emoji when someone posts something genuinely funny. anwyay, i don't take it personally (and i should probably stop posting it on other's posts)...
Kal Madda
macrumors 601
Yeah, sometimes I use the laugh react in place of a disagree reaction, since we don’t get a disagree reaction in normal threads. I probably shouldn’t use it so much though, I’ve been trying to get out of that habit. It’s also hard though when others are laugh reacting at your comments.
Someone posted this link in another discussion; it's worth reading.
tonsky.me
(Only for those who can read a text longer than three lines with examples, the others can contribute with the mocking smiley face that distinguishes them).
It’s hard to justify Tahoe icons
Looking at the first principles of icon design—and how Apple failed to apply all of them in macOS Tahoe
tonsky.me
(Only for those who can read a text longer than three lines with examples, the others can contribute with the mocking smiley face that distinguishes them).
Last edited:
Kal Madda
macrumors 601
It’s funny that you think you know my level of experience. I do have experience with web development, as I said, I just don’t do it for a living… Furthermore, I have friends who do do it for a living, and they agree with me…
I do have experience. Despite your claims otherwise… A well designed page will work with rounded or square corners. This is a basic that’s taught in html and css classes, where we’re taught to design pages with multiple form factors and display sizes/shapes in mind, so it’s surprising even that eludes you…
I can easily see the difference in corner radius… And yes, I understand concentricity…
The border is unnecessary. They should just lose the extra weird border and use a consistent window design regardless of window size, “problem” solved… Oftentimes less is more…
They shouldn’t have counted on the shape of the display or window to always align with in-app elements. Good web design should be dynamic and adaptable for a number of different situations. And window corner radiuses were already different between platforms…
They should have already been considering what their web app would look like with rounded corners, as most web activity is either on smartphones or tablets which nearly all have rounded display corners… If they weren’t thinking about that with their design, than all I can say is they weren’t thinking ahead very well… 🤷🏼♂️
Furthermore, this poses no issue to the app’s functionality…
Wow, that doesn’t come across as rude and dismissive at all… 🙄🤦🏼♂️. Yes, I’m “young” and a freelance designer. That doesn’t mean I “undervalue experience” or haven’t had any mentors or others to bounce things off of. That doesn’t mean I don’t work with other creatives… 🙄. You think you know so much about me, but you don’t…
I’m well aware of the effects of smooth curves in art and visual design, just hadn’t specifically thought about it in context of macOS Tahoe’s new UI being more appealing with more rounded elements.
And there you go again pretending you know my experience or skill level when you don’t… 🙄🤦🏼♂️
Last edited:
kagharaht
macrumors 68020
Just updated and I'm sure you all know how bad Apple Music UI is with Tahoe... What an embarrassment.
star-affinity
macrumors 68020
”Looks fine to me” and shows a screenshot with a very unclear ”Continue” button and very unclear button background for the ”Remind Me Later” button. 🤔