Eshu Marneedi

As the betas progress, my hope and patience dwindle

Take a picture to remember me by. You’ll never hold all the details in your mind.

Hot off the heels of my iOS 26 “hands-on” article in July, my reactions to the new Liquid Glass design were mostly positive. I had written the review largely using the first and second betas, where Liquid Glass tab bars had their more translucent, “glassy” appearance before they were modified in Beta 3. Still, I tried to remain neutral on specific design oddities and nuances because I knew the software would change, and when Apple removed the “glass” from Liquid Glass in Beta 3, my review largely remained unchanged because of how agnostic — or, I should say, future-proof — I wrote it to be. I remember iOS 7 and how much Apple changed the interface in the beta period, so while I left in some quibbles about the Safari contrast and general complaints about translucency and so-called concentricity, I left the specific design criticism to the text-based social networks.

When the Beta 3 shenanigans happened, and I installed it on my device, I had already been working on the review and wasn’t going to rip out the criticisms I had about the translucency because, in the back of my mind, I knew Apple would reverse the changes. They just seemed buggy and out of place, and even though I didn’t like them, I felt that the best outlet to express that wasn’t my long-term review, but some mere complaining on social media. My intuition was right, and Apple did go back to the glassy look of previous betas. But the whole kerfuffle made me look closer at the Liquid Glass situation, especially after reading others’ thoughts on social media. It was particularly a post by Federico Viticci, the editor in chief of MacStories who extensively reported on the iOS 15 Safari design, that brought these criticisms to the front of my mind. In the end, I linked to Viticci’s complaints in my otherwise-positive piece, because this time, I concluded that Apple most likely wouldn’t roll back the changes further.

Viticci’s complaint, in a way, shook me into realizing I was looking at the betas with rose-colored glasses. I had instinctively assumed Apple would tweak the operating systems over the summer and that I wouldn’t have to complain about them, because by the time my critiques had been published, they would be out of date. I was wrong about that — five betas later, Liquid Glass more or less looks identical to the first time it went into beta. Instead of editing my original review, which still remains positive with no asterisks or double daggers, I think it’s clearer (and more honest) to write an addendum. Liquid Glass, as of iOS 26 and macOS 26 Tahoe Beta 5, is far from finished, and I can’t seriously believe Apple intends to ship this software in six weeks when the new iPhones are released. This sense of panic has set in over the past week as I’ve been using Beta 4 and Beta 5, and while I hope I’m wrong, I feel Apple has settled into the beta rut, and we won’t see any concrete changes to the operating systems until iOS 27.

I can no longer retain the sense of neutrality I originally carried in my hands-on review because my sense of optimism has vanished. Apple’s software development timeline is much more distorted than one would assume. As I’m writing this, Apple is probably working on Beta 7 or Beta 8, which usually are the final releases just before the iPhone event. If Apple’s designers wanted to drastically change how the interface looked — a process I think is necessary at this point — they would have done it at least by Beta 5. (For context, Beta 6 is when Apple gutted the old Safari 15 tab bar design on iOS and replaced it with the iOS 18 implementation. It wasn’t perfect, but it was getting there.) iOS 26 Beta 5, however, is sloppy design, and macOS 26 is a heinous atrocity. Unless Apple somehow plans to ship iOS 18 on the new iPhones 17 in the fall, this is a five-alarm fire for Cupertino. The platforms lack the polish expected in a fifth beta. I don’t expect them to be perfect by any means, but they should at least be reliable for developers to build on. I haven’t heard from a single developer confident that they can build on these versions without feeling like they’re working with a moving target.

On iOS, the most prominent concerns remain contrast and legibility. The tab bars in the App Store and Music apps are great examples of how poorly conceived these core tenets of interface design were. When a tab is selected in iOS, it is highlighted in the app’s accent color with a translucent background that attempts to create enough visual separation between the messy content and the colorful icon. This attempt falls flat on its face when that icon’s color matches the background, such as a pink or salmon-colored album in Music or a blue App Store listing — it’s genuinely illegible. I don’t know how anyone at Apple doesn’t see this as a problem. These aren’t premature nitpicks — if a core element of an app’s interface is illegible even 5 percent of the time, that’s a failure in interface design. When core interactions, such as deciding when the tab bar minimizes and expands on scroll, are changing in Beta 5, that’s a failure in interface design. (Apple changed the behavior in Tuesday’s beta; tab bars no longer expand until a user scrolls all the way up to the top, which is boneheaded.) How are developers possibly expected to develop for a platform that has no concrete design philosophy?

As John Gruber, the author of Daring Fireball, said on Mastodon, this is how design critique works. Every time I’ve tried to explain on social media why iOS 26 just doesn’t function well, I’ve been stopped by people who I can only describe as brainless Apple sheeple, usually explaining how a beta should not be criticized even in the slightest¹, as if that’s a sensible retort. This is how design criticism works, and Apple hasn’t been given enough of it this beta cycle. We’re in the fifth iteration of this software, and Apple’s finest interface designers are pumping out icons that look like they’ve been lifted from Windows Vista. Apple’s own SwiftUI apps, like Passwords, still have their navigation titles broken on the iPad. Toolbars on macOS still look as if someone who just got their first Photoshop license began toying around with the drop shadow control. There is no sense of polish to these interfaces, and they’re still littered with scant animations, buggy controls, and a blatant lack of legibility.

When scrolling in an app like Music or Notes — apps with a decent amount of text — the status bar on iOS blends with the text too much, hindering readability. What happened to the safe area? Apple has instructed app developers for years to treat the status bar and home indicator as precious areas where content doesn’t belong, but now, content bleeds past the Dynamic Island and status bar, leading to some of the most illegible text in the entire operating system. And despite Apple’s developer documentation’s continuous reminder to use tinted Liquid Glass for standout app elements, Apple seldom uses it in system apps, instead opting for the iOS 18-esque tinted controls. Part of the reason is that there’s no good way to use them in toolbars — the tools for designing interfaces like that don’t exist without hacky workarounds. (In SwiftUI, toolbar items with text can’t use tinted Liquid Glass.)

While Apple has mostly addressed my woes about Safari tab bar selection on macOS — and the relative jank of the Show Color in Tab Bar setting — these changes haven’t been transplanted to the iPadOS version of the browser. Merlin Mann, a podcaster and writer, also screenshotted some examples of Safari in macOS Tahoe not working as expected, and his example is particularly bleak: selected tabs and background tabs have next to no difference in accent color. This is a table-stakes interaction in any macOS and iPadOS app, and Apple hasn’t been able to get it to work with any decency five betas in. Sidebars in macOS still make little logical sense: They appear as if they’re floating atop the primary window’s content, yet they let a smidgen of the desktop wallpaper’s color through (à la macOS 10.10 Yosemite and beyond). Where is the color coming from if the sidebar is layered atop the otherwise opaque window? Users aren’t likely to notice this level of detail when they’re using their computer, but they will once their apps mirror their content behind the sidebar as Apple encourages developers to do so.

This nonsense — which carries over to the indescribably putrid toolbars in macOS Tahoe — was perfectly described by Jason Snell, the editor in chief of Six Colors, in his hands-on impressions: “…it feels like Apple has lost its balance in a quixotic attempt to make every app look like a photo editor.” macOS, much like the unreadable tab bars of iOS 26, forces tab bars to blend in with content, which works great in apps where immersiveness is encouraged — like photo editors — but is otherwise illogical (or “quixotic”; I love Snell’s choice of vocabulary here) in any other app. It really became clear to me how far macOS has lost its sense of individuality when I scrolled past an iPadOS 26 screenshot from Steve Troughton-Smith, a developer, which I initially thought was from macOS until I read his caption. With the addition of the menu bar and the new shared design idiosyncrasies between iPadOS and macOS, some apps are quite literally indiscernibly similar across platforms. That’s not a negative on iPadOS, but it is on the Mac, since no Mac has a touchscreen that would require interface elements to be so far apart. Yet, alas, they are.

This article sounds like a rambling rant, because it largely is, and that’s by intention. My rosy, optimistic thoughts about Liquid Glass and my gush on how stunning it is are available on this website, just a few posts down, for everyone to read. But just as I gave Apple positive feedback a few weeks ago for its design work, I also think it’s in the company’s best interests to take negative feedback to heart, too. I’m not asking for a Beta 3-style rollback of Liquid Glass, and I still find that release too extreme. I don’t even particularly prefer it over the current iteration, which is to say, I hope neither ships to general consumers in the fall. I feel bad that I don’t have a checklist for Apple’s designers and engineers, too, but that’s just my Apple fandom kicking in again. Why should I, some lowly blogger, provide professional-grade design advice to a company worth $3 trillion? Its engineers, the same ones who made the iPhone X’s gestural interface and the Dynamic Island, should be able to figure this out. While I have faith in their talents, I don’t carry that optimism to their ability to do it quickly enough.

Five betas later, the Mail app on iOS just pulled the Select button out of a context menu for easy access, only to use an X glyph for its Dismiss state, which, at first, I thought deleted the selected emails instead of merely exiting the selection menu. I’m a software developer who has religiously studied Apple’s Human Interface Guidelines, and even I, a person who knows that wouldn’t be an acceptable pattern in Apple design, got hung up on that detail when trying out the button for the first time. How is a run-of-the-mill iPhone user expected to intuit that? Whatever happened to labeling buttons with text that describes their function? I understand that such concepts would be unfathomable to Apple’s glass-enclosed designers with clean slate white countertops and oak tables, but for the rest of us who live in normal homes, text labels are often handy in software interfaces. Seriously, who thought text labels for Done and Dismiss buttons were too cluttered?

If it took five betas, or two months, for Apple to add a Select button to the Mail app, only for it to be so haphazardly designed, how long will it take for major wrinkles like tab bar and toolbar selection to be ironed out? Maybe all of these quibbles will magically disappear in the next beta, and Apple’s platforms will be moderately usable again, but what rationale has Apple given its beta testers and developers to believe that? These aren’t typical beta bugs (“Messages crashes upon sending a GIF”); they’re specific, detrimental usability quirks found throughout all of Apple’s latest platforms. I don’t think staying silent and letting out a few prayers is an actionable solution to a host of issues that will hit millions of people in a little over a month — this is how design criticism works. I don’t think it’s unreasonable for me to ask some of the finest user interface designers in the world for a tab bar that lets me read the selected tab’s title.

This reproach mostly serves as an epilogue to my otherwise positive Liquid Glass review, but it reflects my current state of emotion toward the update: hopeless. The very last conclusion anyone, especially Apple, should take from this piece is that I somehow hate Liquid Glass or wish for the changes to be reversed. I think it requires and, importantly, deserves work to succeed. In its current state, Apple would be reckless to ship it to millions of iPhone buyers in the fall, and I think that ought to be pointed out before we’re past the point of no return. When seasoned, platform-native developers complain that they’re unable to figure out how to proceed with their redesigned apps this year, how are large development teams from Fortune 500 companies expected to? iOS 26 is unpredictable, unreliable, and half-baked. macOS 26 is a national embarrassment beyond words, so much so that I think it is irredeemable. I don’t write these words lightly — I write them out of months of hope that Apple would right its wrongs and craft an elegant solution. As the pages disappear, slowly floating off into another year², my hope dwindles, and so does my faith in Apple’s agility.