Hacker News new | past | comments | ask | show | jobs | submit login
What’s New in macOS Big Sur: Human Interface Guidelines (developer.apple.com)
408 points by tomduncalf 23 days ago | hide | past | favorite | 366 comments

I have to be honest and go completely against the crowd here - but I think the UI updates are very good. They make the two OSes look cohesive but in a tasteful way. The extra padding in all of the UI is the only thing that I chafe at a little bit, it feels unnecessary. I really don't think they're going to put out touch screen Macs.

So I'm not sure what there is to get upset about. I don't see any big functional changes in the OS aside from reworking the notification center UI a little- and I like this new version more. It's already been confirmed that they aren't locking down apps to the app store or anything like that.

It seems like people think that because they're taking design cues from iOS they intend to lock it down like iOS and they're transferring that fear into dislike of the new UI itself.

Apple knows that a large percentage of the people who buy its computers are software engineers and if they ever do fully lock their computers down in that way they'd be ceding that marketshare entirely, along with catching quite a bit of heat for it in the press. Maybe eventually that calculus is worth it to them but I don't think that's what this is a step towards. I think they just want users who are fully plugged into their ecosystem to feel like it's a cohesive experience across platforms and this achieves that.

Well that and the fact that tech website users basically complain about every UI change. Hacker News is full of people who think XFCE is the height of desktop experience. This is fundamentally different than apple’s core fan base who seem to like any change just for the sake of change.

I agree, any change will be frowned upon. It just annoys me that I feel like they're drawing the completely wrong conclusions.

Apple was the company that dropped the headphone jack from their phone, shipped a laptop with only USB-C ports etc... they don't shy away from making big controversial decisions if they feel like it's good for them.

If Apple was going to do this to MacOS, they'd have already done it. They wouldn't be spending years tip toeing towards it like people seem to think.

Instead, they've rolled back some really unpopular changes like the crappy keyboard in their laptop, they just shipped a new Mac Pro that is clearly designed for professionals. This chip transition will result in Apple's laptops having better battery life, specs and heating than any other laptops available on the market. It won't even be close.

Additionally, they're actively trying to bring a huge swath of developers who currently only develop for their mobile devices into developing for the Mac. If they were trying to turn Macs into something for college students to watch Netflix on they wouldn't be doing any of this.

It's insane to me that people can see a UI refresh as a sign of doom while ignoring that Apple has actually been investing in the platform quite a bit more than they were a few years ago.

I still don't understand how anyone with hundreds of billions of dollars ships really crappy anything. Like do hundreds of people just pretend that a crappy keyboard isn't garbage or is it so siloed that most people never get an opportunity to register an opinion?

"Hey dev, you don't like that new keyboard pushed by your director/VP? Here's your termination notice!"

"We love our developers and pride ourselves on individual expression and transparent feedback."

-- Corporation 101

This is part of it, but I imagine there was a lot of sunk cost in there as well.

Imagine: you're in charge of Apple's new keyboard. You spend a bunch of time and money making the thing as thin as possible, because that's the overall design trajectory.

Of course, you test this thing internally, and it's okay. It takes a little while to get used to, but isn't awful to type on, and a few months of stress testing seem to indicate that this thing will work in the real world with an acceptable defect rate.

Besides, your management chain loves this thing, because It's Thinner(tm).

And then, you ship it, and discover that this thing breaks way more often in the real world.

But, because Apple, you're a company that never publicly admits to doing anything wrong (because lawsuits), which takes customer feedback with an overly-critical eye (philistines must be using it wrong), and which is more likely to spend it's entire capitalization on cat treats than it is to interrupt an ongoing hardware design cycle.

It's not just management -- it's the company culture.

On the one hand, this enables them to do bold things and endure incredible criticism before being proven right. The iPod, no optical disks, bundled GPUs, AirPods, etc.

On the other hand, this enables them to do bold things and endure incredible criticism before being shown horribly, horribly wrong. The snowflake keyboard, the touchbar, the iPhone antenna, etc.

I also suspect this is why Apple's hardware design is just so overall excellent. They identify perfection, and relentlessly pursue it. Sometimes Apple gets it wrong, and it shows, but when they get it right, they get it really right.

>On the one hand, this enables them to do bold things and endure incredible criticism before being proven right. The iPod, no optical disks, bundled GPUs, AirPods, etc.

When the iPad first came out: "OMG, you can't print! How insane." Probably forgetting other things.

Apple does a ton of things wrong and arguably forces some changes before a lot of the world is ready for them. But they've never prioritized supporting legacy and, in fact, arguably have had a pretty consistent strategy of eliminating legacy at least one or two beats early relative to other companies. Certainly I've found some changes too aggressive. OTOH, there are usually ways to adapt that I've found no more than mildly annoying at first.

> When the iPad first came out: "OMG, you can't print! How insane." Probably forgetting other things.

I very much don't recall anyone complaining about printer support. The primary jab at launch was "it's just a big iPhone". Which was in fact true—it just turned out that "a big iPhone" was pretty damn good, especially when actual iPhones had 3.5" screens.

I'd say most of these sorts of complaints are basically legitimate. When the iPhone came out, it didn't have copy and paste, and that was in fact extremely annoying!

I'm not going to spend the time to look but there were definitely complaints about printing which was still much more of a thing than it is today. (It's still sometimes annoying with both iPad and Chromebook but it's occasional enough that I just mail what I need printed to a computer.)

The thing for me with the original iPad was coming to the realization that it was a (very nice) luxury device that was mostly useful for me (at the time) when traveling. And, when I gave one to my dad, it was a revelation relative to computers that he never could really use.

The keyboard isn't that bad. It had reliability issues which suggest a failure of their stress testing, but it wasn't somehow obviously broken when it shipped

> The keyboard isn't that bad. It had reliability issues which suggest a failure of their stress testing, but it wasn't somehow obviously broken when it shipped

I had mine replaced twice, but the third revision (which according to Apple is an improved version) has been reliable (except for the key labels wearing off) though I still greatly prefer the 2012 design.

My biggest issue now is the trackpad and its incredibly broken palm rejection that causes the cursor to jump all over the screen. This isn't a settings issue (settings do not help at all) and it was never a problem on older Apple laptops or on the iPad, which has a huge touchscreen with flawless palm rejection. Unfortunately (for me and for Apple), wearing an Apple watch seems to make the broken palm rejection and random cursor jumping even worse.

The 2019 MacBook keyboards are ok. The ones before were worthless. The ones in 2005 had jitter resulting pro keyboards in double key presses rendering them also total garbage.

I am typing on a 2017 MBP keyboard that never broke. While I generally share the criticism it is neither garbage nor worthless. These unnecessary juvenile hyperbolisms are so boring.

The keyboards had a higher fault rate and many other issues and Apple has corrected its mistake, albeit 1-2 years too late from a consumer perspective. That's all there is to the story and the only semi-relevant competition to MBPs was STILL just ThinkPads and XPS, even with the main input device being significantly worse than previous iterations.

> These unnecessary juvenile hyperbolisms are so boring.

If you used the 2016 keyboard, you wont say that.

I eventually threw away my apple keyboards from 2005 so they are literally garbage.

Regarding the 2017 keyboards, I'm not even talking about the fault rate. They had no escape key and the key travel was too small so it was like typing on a touch screen.

> and the key travel was too small so it was like typing on a touch screen

No, no it's not really.

It's all about preference and tolerance. I bought my Mum an 'old' Apple USB keyboard (i.e. similar to the pre-2017 laptop keys) when her PC's existing KB broke, as I liked mine very much. Being a life-long touch-typer, she just couldn't get on with it, because the travel was too little for her. She's now very happy with a KB with mechanical (Cherry) key switches.

In contrast, I prefer the older Apple KBs, but can get on perfectly well on my 2017 MacBook Pro's KB; and it's not like a touch-screen at all.

>No, no it's not really.

Then maybe I only used broken ones with no perceptible key travel whatsoever.

I wonder, how my use of my 2017 MBP differs from yours - I am constantly burdened with key presses that are registered twice or not at all, keys getting stuck. A bit of dust getting in the switches sometimes renders some keys unmovable - the switch will actuate but there will be almost no perceptible key travel.

I agree that when it does work, there is _some_ feedback from keypresses, but seriously, if they can fake the clickability of the touchpad via haptics, why couldn't they do the same for the keyboard? Out of all devices that I develop on, the macbook is the most frustrating, mostly because of the keyboard.

Hmm, the travel was really something I got used to quickly, although I also prefer more travel. I found the noise and the lack of an escape key much worse and these issues have recently been corrected.

Still, I don't think these issues made the KBs true garbage. Such examples would be nearly any non-MacBook touch pad or often enough display hinges. Far worse than a slightly less convenient keyboard which can always be supplemented with an external bluetooth keyboard as well. Can't really do that with hinges.

It fails within a few months of usage. Even when replaced twice

And replacing is even more of a problem as it's not simple or cheap. And wasteful.

It’s almost like the amount of money a company has doesn’t correlate with its ability to make a great product :)

I am mostly upset about the low contrast. I hope that gets improved before release.

And sad about the change in sheets and alerts... I’ll probably get over that.

This is a bit off topic but in regards to contrast, I’ve been using Apple’s Music program lately and I absolutely cannot tell the difference between shuffle/repeat on or off. It is the lowest amount of contrast I have ever seen on a toggled button.

I figure it must be something with my machine but haven’t really delved too deeply and couldn’t find much when I looked.

Was hoping for a bit more out of Music. It’s decent, and certainly better than what iTunes had become. But sometimes I’m left wanting.

Although having my Pops be able to share a playlist over iMessage with me was a stellar experience in terms of how seamless it was and having it sync from my phone to my computer.

I don't think it's your machine, although I hadn't tested this until now because I almost never use shuffle. :)

I think the problem is that they're indicating state by toggling between your chosen accent color and black. If your accent color is, say, pink, the difference is pretty obvious, but if your accent color is relatively dark, like blue or purple, it's harder to see. It seems like they could solve this by toggling between the accent color and gray, or "lighting up" a solid roundrect of the accent color around the control when it's on.

Thank you! I just changed my accent colour to green and now it’s easy to tell the difference.

That may be a wonderful clue, thank you :-).

Where do you set accent colour?

It's in the "General" section of the System Preferences

The contrast changes seem varied. For some features they've increased contrast: "windows" are now white with black-ish icons. You can't get a higher contrast ratio. Other features have received the translucency treatment, and seem to have lower contrast.

I'm excited to try it all out, but also unsure how comfortable I'm gonna find it. The large white expanses seem glaringly over-bright to me, at first glance.

The toolbar used to be grey and is now white like the content. And the contrast of text on the translucent items... But overall the style is very consistent.

If it is bright in light mode, it might be nice and dark on dark mode. We’ll find out once people start running the preview

While not an excuse, the design looks like it has much more contrast than 10.10 - 10.15.

I don't see it as a sign of "doom," but rather more regression. It took YEARS for Apple to admit that their keyboards were shit, and they still haven't admitted that the emoji bar on their so-called "pro" computers is an embarrassing and failed gimmick. In fact, they've doubled down on it and now don't offer ANY proper laptop keyboard.

But that's off-topic to some extent; I scanned through this page looking for some good news for developers. Topping my list is support for SVG assets. NOPE. WTF? Android has had this how long? Apple's "@2x", "@3X" bitmaps were amateur-hour to begin with, but a decade later? No vector support?

Come on.

And still no user-definable color schemes. Windows (and Unix GUIs, for that matter) had them for decades, while Apple forced its late-'80s inverse scheme on every user that whole time. The hard-coded "dark" theme is a half-baked, belated admission that staring into a glaring light bulb with black text on its surface all day may not be the best way to work.

This would also have been a great time to abandon the single menu bar, but nope; that poor decision continues to plague Mac users. Really, this should have been canned in the transition to OS X, but now there's even less of an excuse.

I'm sure there will be some good changes, but from this list I'm not encouraged. Getting rid of the distinction between toolbar and app content... WHY? The "flat UI" experiment was a failure. Let's continue to take tasteful steps back toward GUIs that communicate in universal visual terms whenever possible.

> Topping my list is support for SVG assets. NOPE. WTF?

From the Xcode 12 Beta Release Notes:

> Added support for Scalable Vector Graphic (SVG) image assets. These preserve their vector representation with deployment targets of macOS 10.15 or later, iOS 13 or later, and iPadOS 13 or later. (18389814)

> Android has had this how long? Apple's "@2x", "@3X" bitmaps were amateur-hour to begin with, but a decade later? No vector support?

PDF assets have been supported on all Apple platforms for many years.

The announcements today represent years of work by thousands of engineers and designers. It’s impossible to cover everything in a 2 hour keynote, let alone one geared towards the media/general public. Even the 2 hour “State of the Union” that’s geared towards developers barely scratches the surface, let alone a single webpage.

@2 @3 allows you to remove detail when needed

PDF !!!= SVG

But it is vector.

I know. IMHO, it's nitpicking on a technicality. Use cases are radically different.

> But that's off-topic to some extent; I scanned through this page looking for some good news for developers. Topping my list is support for SVG assets. NOPE. WTF? Android has had this how long? Apple's "@2x", "@3X" bitmaps were amateur-hour to begin with, but a decade later? No vector support?

Not that I don't think they should support it (they should), but this is almost certainly due to perfectionism and bias, not amateurishness. Apple has an institutional focus on good iconography, and good iconography almost always has different representations for different sizes. The icons scale within a certain size range, then switch to a representation more appropriate for the size. I would be shocked if this is ever completely eliminated from Apple's DNA.

They tried non integer scaling, and it just didn’t work out. You could enable it with some “default write” on 10.4-10.6 IIRC and you could witness the various iterations, but basically the result proved that pragmatically integer scaling gave much more consistent and predictable results.

And yet it works pretty well on Win10, judging by my devices (that have scale settings ranging from 150% to 225%).

> This would also have been a great time to abandon the single menu bar, but nope; that poor decision continues to plague Mac users. Really, this should have been canned in the transition to OS X, but now there's even less of an excuse.

If you are using a precise mouse or trackpad, a menu bar at the top (or edge) of the screen is much easier to than a menu bar anywhere else on the screen, because it has "infinite" extent. This has been known for decades, and is why macOS places the menu bar at the top of the screen and the dock at the bottom or edge.

However, Apple has made remarkable progress rethinking pointer design on the iPad, as seen in this excellent video, "Build for the iPadOS Pointer":


I still don't have any USB-C devices or wireless headphones. May be in 10 years I'll change my opinion, but as for now, I still think that it was wrong to ditch USB-A and headphone jack.

> Hacker News is full of people who think XFCE is the height of desktop experience. This is fundamentally different than apple’s core fan base who seem to like any change just for the sake of change.

that statement seems perhaps too harsh to both HN crowd and Apple's core fan base, I spend more time on HN than I care to admit and I can assure you that I do not like XFCE desktop experience at all, my desktop is running Debian 10 with Cinnamon on it, but UX is not something that I'd ever list as one of top 5 features of this setup. Similarly I've been using Apple laptops for about 14 years now and do consider myself a part of "core fan base", for the record I hate the butterfly keyboard, despise touchbar with a passion, and am utterly disgusted by the decision to remove MagSafe chargers.

To quote great Dr Tobias Funke - there are dozens of us.

Nah, I watch most of the WM-related threads because it's more interesting to me than I am willing to admit in public life. XFCE gets trotted out generally any time GNOME, KDE, MATE or Cinnamon does, but especially in the case of the former two. There's usually a cadre of MATE and Cinnamon reps as well, of which you might as well consider yourself at least an honorary member at this point.

StumpWM, i3 and dwm are also commonly brought up contenders as peak desktop Linux.

If it doesn't apply to you, it doesn't apply to you.

Maybe i3 wouldn't be peak desktop Linux if other environments realized that virtual desktops extending across all monitors is bad design.

I was perhaps a bit unclear in my post - I use Cinnamon and am thankful to those who made it, however I don't think that or XFCE or MATE or KDE are the "height of desktop experience", they get the job done.

In my opinion height of desktop experience was Mac OS X 10.6 Snow Leopard - I still miss workspaces, and TotalSpaces2 is a valiant effort but less likely to work with each new update.

I haven’t seen an actual complaint about what’s wrong with XFCE.

Simple minimalism is certainly preferable to whatever macOS can be described as by many, sure. I don’t see a problem with that.

I feel exposed :D. I really enjoy using XFCE and run it on all my machines.

You should feel proud. I pull in some XFCE packages to compliment my i3 desktop and it's absolutely great.

MATE and LXQt are two other comparably simple desktop environments that I like.

I get comments from my bewildered MacOS-using friends when they see that my 400$ Linux machine is faster than anything they've ever tried.

Where should I start in order to learn this superpower?

Why are they hating on or even mentioning XFCE? XFCE can both act and look like Berdoo / San Bernadino, the last OS11 release, what's the problem with it?

You new-fangled San Berdoo OSXI users get off my lawn.

Cause they're going to be $5,000 poorer the next time they buy a laptop.

At least we don't have to manually adjust fan speed every time WLAN resets after sleep <3

I agree. I can run XFCE on a modest set up, and it does what I want.

Some UI changes make sense. You wouldn't want to run XFCE on a touch-capable device, whereas GNOME3 is perfect for that use case. But the macOS UX changes make no sense to me. And really, even OS X itself has always been quite confusing, compared to its more traditional alternatives. Now we're finally on Mac OS 11, but that hasn't changed either.

> Well that and the fact that tech website users basically complain about every UI change. Hacker News is full of people who think XFCE is the height of desktop experience.

.... Yeah, that's pretty much true, and for very good reason. Every time an interface changes, it forces people to stop using the system and figure out the new interface. As such, while it is possible to make changes that are beneficial enough to justify the cost, UI changes are very expensive and you'd better have a very good reason to justify breaking everyone's workflow. And by that standard, yes; XFCE is nearly perfect. It's been on the 4.x series since 2003, and going by screenshots I strongly suspect that even a user from XFCE 3.x would find the very latest version to be a drop-in replacement. And it does have decent quality, which combined with its general "get out of the way and let me work" vibe and 20 years of continuity makes it easily one of the best options available.

It doesn't look like macOS 11's new UI is going to need macOS 10 users to "stop and figure out" the interface in the way switching from macOS 10 to Xfce (or vice-versa) would. The buttons aren't all moving around to strange places. The functionality isn't materially changing. What they're changing is the design language, not the fundamental UX.

Well, as an occasional MacOS user, it still pisses me off when I discover that maximize button is in fact full screen button.

As far as I know, that part of the UX will remain consistent with macOS Big Sur, and it'll still piss you off. :)

(You can hold down Option and use the green button as its old "Zoom" function, but I think there should be a system preference to reverse that so it's zoom by default and you hold down Option to go into full screen.)

Yeah, except it is NOT a "maximize" button and never was. macOS just doesn't do maximized apps the way Windows and some Linux DEs do.

Before Yosemite the green button worked like a maximize button, and you had a separate fullscreen button on the other side. Now you need to ⌥-click the green button to maximize.

No, that's not what it does on macOS. It simply makes the window larger and has done that ever since it was introduced, I believe. (I think for many document-based apps this just means to make the window taller, for example.)

IIRC the green button was "zoom", meaning it should adapt the window size to the window content, which in most of the cases resulted with maximizing.

And it's still fundamentally flawed as ever. (since the suggested behavior in the HIG presumes contextual intent is something that's generally clear or even decidable from the content.)

I know, but it as close to maximize as it gets. That's not my point, but that the behavior has changed dramatically.

I've been a full-time Mac user for nearly 20 years and I'm kind of with jablan here. :)

I don't know about you, but my brain is pretty plastic. My workflow isn't so sacred that I'm not at least willing to consider that there are other designs that I may find enjoyable or even more efficient to use.

I don't think for some it is about plasticity, but really to embrace change on one's own schedule. That's my preference and it's a little piece of freedom. I did this recently moving from GNOME to XFCE, to experiment with what XFCE offers specifically with memory consumption, usability, etc. and after some bumps, I am rather happy.

Optimizing things where the primary focus of attention is what I am looking at mostly is more relevant to me than a shifting design languages. But, I know others will have different views on this and this is based on what people are actually happy with.

XFCE is the ehight of desktop experience.

>fan base who seem to like any change just for the sake of change

I think it's more that (visual) UI Design has the same cyclical property of most other "visual design" like clothing fashion, interior design,... Things look fresh when they're new, after a few years people get bored of the look, and what was once considered modern now looks dated.

Especially for a company like Apple that has built part of its reputation on having the best UI, they can't afford looking behind on the competition. Calling it "change for the sake of change" is partly correct if we were purely talking about products, but it ignores the fact that Apple is as much a marketing company as it is a product company.

Aside from that I think there's also a strong functional argument to be made. iOS and iPad share the same OS, so it makes sense for those to be consistent. As the iPad has matured it's approaching desktop levels of productivity. So now we're in a situation where the iPad is becoming like a Macbook without the keyboard. With so much overlap in functionality, it also makes more sense for those 2 experience to converge rather than diverge. So we're hearing complaints about the "iOSification of the macbook" but they are ignoring the missing link; iOS = iPadOS = macOS

What's wrong with xfce?

Nothing. It’s just not “the height of desktop experience,” is all.

Aside from the default theme (which can easily be changed), what do you think it's lacking?

Nothing. It's just that calling it "the height of desktop experience" would sound weird to way too many people. (Some, for example, think it's emacs.)

What's the height of the desktop experience?

It's a deep question. Some think that it is a bunch of virtual terminal windows (the idea driven to the extreme in the GUI of Plan 9), while some would feel at home with a desktop being kind of a picture of a real desk. Still others would prefer a desktop that consistently follows an object-oriented pattern: you click on an icon and get a list of the object's properties and methods. (Unfortunately, too many screenshots that are meant to showcase desktops that people like just show a flowery background image and a bunch of random windows on top of it containing a picture of a cat or some manga drawing.)

Windows 2000 or KDE 3.5

Windows XP? Or 7?

Both have a classic theme.

Well it’s close but still not as good as Win2k was.

"XFCE" is a funny way of spelling "i3wm."

> Hacker News is full of people who think XFCE is the height of desktop experience.

There are those of us who will have to choose a new DE because Big Sur apparently won't run on our perfectly capable pre-2013 MBPs, and have to choose the least clunky. In that group, Xfce is indeed the bees knees.

XFCE is good yes. I don't understand why anyone has problems with it. It's perhaps projecting. They're projecting. Wishing their San Bernadino OS Eleven looked as good as XFCE. It takes nine minutes to set up an XFCE desktop looking better than OS XI San Bernadino. So I mean, obviously they're retro-projecting or whatever we'd call it. XFCE for life.

XFCE is obviously not the height, but it is very well executed

I'm more likely to prefer XFCE than Gnome, for example

That being said, I also use and like desktop Apple products so there's that

I agree, I think the new UI looks great.

People are taking the UI / ARM change as apple taking away more control, but macOS is THE development platform of the entire apple ecosystem.

So yeah, the ecosystem may be tighter knitted, but they sure as hell will not lock down macOS like they do iOS, they have so many mac users who are developers and apple are absolutely aware of that.

People complain about features in catalina and other recent versions of macOS 'locking things down' - it's incredibly easy to bypass the stuff that makes a difference using the terminal.

In a world where every company in the world wants to track and harvest your data, we've somehow got to the point where people bash apple for actually taking privacy and security seriously.

Apple locks down MacOS more and more with each major release. Some of it seems well justified but the totality of it is concerning to me. Being able to defeat some of it in the console doesn’t change what is an increasingly abusive relationship. I feel like we’re rapidly approaching a point where Apple enforces the same strict control over MacOS that they do over the iOS App Store and they’re softening people up towards that in baby steps with each incremental restriction they implement.

Maybe I’m just paranoid, but I personally decided to stop buying their computers because of it.

Cory Doctorow talked about these worries quite eloquently back in his 2011 keynote to the Chaos Computer Congress, titled "The Coming War on General Purpose Computation".


I mean, that was nearly a decade ago and we're still doing just fine for general purpose computation.

At what point do we stop saying "he was so prescient, look we're almost there," and just admit things haven't turned out as badly as people feared? DRM has turned out be more of a nuisance than an apocalypse, and general purpose computing is at least as available today as it was in 2011.

In some ways it is even easier; for example anyone with an email address can get access to a cloud-hosted computer for free.

Exact! Or build your own Kubernetes cluster at home on Raspberry Pi for less than the price of a phone or TV.

Depending on measures taken a small kitchen fire can be a minor annoyance or burn up the family home and the family dog with it. I don't think there is any doubt that there have been elements that had there been no negative reaction would have pushed for much more oem control of user machines nor is there any doubt we are in fact moving progressively towards more locked down machines.

Finding parties that would benefit is a trivial exercise one has only to look towards app store revenue, Intels idea of selling "upgrades" to your cpu via software unlocks, or Verizons ability to rent you access to your gps for $10 a month.

Others have been pouring water on a fire that would have already filled your kitchen with smoke by now otherwise and you have merely failed to attend to this fact.

I disagree, I don't think they're anywhere near locking down to the iOS level.

Could you explain some of these incremental lock downs? I honestly can't think of any that are genuine restrictions, but that might just be because I don't encounter them with what I do.

How about the non-modifiable hidden system partition? Inability to create directories under /

That alone has stopped me moving my Macs to the latest OS after testing.

I'm genuinely curious: why do you want to create new directories at the root level? Is this something you're just used to doing and it breaks your customary way of working? Is it a "general principle" thing (which I suspect is a lot of the objection the HN crowd really has at the root, no pun intended)? I've been using OS X-based Macs since 2002 and used Linux and FreeBSD for a decade before that, and I don't recall ever having an actual need to create a root-level directory.

> why do you want to create new directories at the root level?

Why not? It's his system, he should be able to do whatever he likes. On most Linux I can rm -rf / all day long and nobody cares. What's the problem with creating root level directories?

I didn't say there was a problem with it; I just asked what the reason for wanting it was.

> On most Linux I can rm -rf / all day long and nobody cares.

So let me turn this around: Why do you think a system that doesn't make it difficult to completely destroy it with one command is a bad idea? "It's my system, I should be able to do whatever I like" is an answer, to be sure.

(N.B.: in fact, "rm -rf /" won't work without "--no-preserve-root" in most Linux distributions, AFAIK, and I presume sudo.)

DOS survived perfectly fine with deltree, whereupon it was possible to delete your file system.

The decision on which bits of the system I can destroy should be mine, not theirs. That's the issue. At the moment, I am getting pushed out of my machine, bit by bit. Every binary I run is getting report to Apple etc.

It is also possible to destroy the system by dropping your MacBook from a very great height. Should it be padded to cope with this eventuality?

> Why do you think a system that doesn't make it difficult to completely destroy it with one command is a bad idea?

This does not seem very charitable.

It is fine if the system warns you about doing potentially dangerous operations (which --no-preserve-root amounts to) but it must not prevent you from ignoring that warning.

Well I make all my directories in C:/ or / just because the paths are shorter to type. That's one reason.

I put all developer libraries that I do not want to put into a system path into /DevelopLibs. On Windows I put them in C:\DeveloperLibs. That way I can build supporting libraries of different versions software under there, eg. /DeveloperLibs/abc-1, /DeveloperLibs/abc-2 and test that I have a self-contained executable that runs on my development system without relying on anything in $PATH.

It was a habit on macOS since /Developer used to be where xcode lived until they moved it to /Applications.

It also makes it easy to look in the /DeveloperLibs directory and see if I've got all libraries I need on this machine (I have different development machines under different OSes).

Removing ability to create directories under the root seems draconian for no good reason, other than the iOS-ification of my machine. Not Apple's machine - my machine.

I don't understand why people would ask "why would you want to do that?" because everybody used to defend the ability to only resize windows from the bottom right corner under macOS until a few versions ago, when it became obvious that resizing a window from any edge was a good idea and always had been. But people oddly defended it for decades before that.

What if they now remove the ability to change your desktop wallpaper? Would we all say "but why would you want to change your wallpaper?"? We seem to accept the gradual lock-down of our own machines. Apple users (and I am one) seem to be the most defensive of the company's behaviours and it doesn't make sense - we just have to accept it.

If Microsoft removed the ability to create a directory under C:\, what would the result be? Would we be asking "but why would you want to create a directory under C:\?"? I think not. It'd be the apocalypse.

As happymellon noted, I just use /opt when I need to.

There are certainly many decisions Apple makes that I don't like and I am always keeping my eye on Linux in case I really feel the need to pull the rip cord. But for me, this is just not a hill of sufficient size to die on. "Apple now mounts the system partition as read-only" just doesn't feel like a slippery slope that leads inexorably to... well, I'm not really sure where people imagine it's going. "This is something I used to be able to do and now I can't and soon that means I won't be able to open a shell!" I mean, maybe, but... probably not. If I had to choose whether "Macs in five years will only be able to install programs from the Mac App Store" or "iPads in five years will be able to install programs from places other than the iOS App Store" is more likely, I'd honestly go with the latter.

> I don't understand why people would ask "why would you want to do that?"

I guess I don't understand why asking the question is controversial; I'm not trying to lead anyone into a rhetorical trap that forces them to admit only axe murderers would ever use this functionality. :) It's simply that in what I'm pretty sure is over a quarter-century of Unix use on my part I don't recall ever putting anything in the root directory of any of my systems except by mistake.

As for "defending the ability to only resize windows from the bottom right corner until a few versions ago," your memory of this is different from mine, I suppose. I recall a lot of Mac users thinking that Apple's one-corner-only approach was a bad leftover from the pre-OS X days that needed to change!

Yeah, TBH I think that more file systems should be mounted as RO by default, and only moved to RW during an operation then restored to RO post system activity.

It would get rid of shitty Mac application documentation that rather than installing itself in correct locations, has troubleshooting steps like:

> Just chown your entire `/` to your current user and `chmod +w *`...

Until Atomic installations become the norm like SilverBlue its probably a good step.

So two things here. The first is that it sounds like you want to be using `/opt`.

The second, is that I am able to create root folders, I just have to remount as rw before creating it and after that I can mount any data folder to `/foo`. I'm not entirely clear why not having `/` as rw by default is such an issue, but you are cetainly not prevented from doing it if you wanted to.

I put symlinks in the root directory for my commonly used paths. Or I did before Catalina. Yes it’s kind of a dumb thing but it also seems dumb and arbitrary that I’m no longer _allowed_ to.

> I'm genuinely curious: why do you want to create new directories at the root level?

Why would Apple want to restrict access to certain memory locations?

Turn off System Integrity Protection. Not a good idea but it will let you make unnecessary changes to /.

All they really did was make it harder to thoughtlessly do. If you can't cope with the procedure needed to make changes to the write-protected folders, maybe whatever it is you wanted to write there isn't actually that important.

That doesn't make sense. What is in / that is dangerous?

Finder hides all of /etc and /bin and /usr for the "protection" of normal users, just like the default view for Windows Explorer under C:\Windows\System. But you can still change the contents on Windows, or see the directories in Terminal on macOS.

Should they hide those directories when you use Terminal too? After all, /etc is important. Your reasoning would also stand up there - perhaps whatever files you were attempting to modify under /etc wasn't actually that important.

When does the "you cannot easily modify files on your own machine" stop?

Would you also accept the inability to change your wallpaper with such glee? Perhaps the wallpaper you wanted wasn't actually that important.

> What is in / that is dangerous?

...it's...the root of your entire system? Are you really asking me why `rm -rf /` is a meme, or are you just being contrarian?

> Finder hides all of /etc and /bin and /usr for the "protection" of normal users

And you can change Finder's preferences to show those folders, like any actual macOS user that's qualified to be making changes to those folders should know. If you lack both the curiosity and the ability to look up how to enable that option, then you frankly have no business modifying them yourself. If whatever you wanted to copy or edit there wasn't important enough for you to take ten seconds to look up how to do it and then follow through with the instructions, then why on earth would you be complaining about it to anybody?

I'm not being contrarian, I am asking why I wouldn't be able to modify my filesystem. It's quite simple.

I can also do rm -rf ~/Documents or rm -rf /Applications or rm -rf /Library which has a similar devastating effect to my life/machine - should we block that too?

I can also fling a hammer into the screen and break the screen - should we put extra thick glass on the screen? I guess I have no business using a MacBook if it is possible to break the screen.

I can also pour liquid into the keyboard - should the keyboards be waterproof? I guess I am not qualified to use a Macbook and have no business using one if I spill a drink.

Using the keyboard is dangerous too because someone can type dangerous commands. Let's stop people using it. Only those with sufficient "curiosity" and those "qualified" should use it, right? How do I get "qualified"?

I am saying that as it is my machine, I get to decide which parts are dangerous. But I no longer can because someone the other side of the world has decided that I am cannot create a directory under /.

I do not ever modify /etc using Finder. Should I only be allowed to edit files under there using Finder or is Terminal permitted?

I guess if we follow this "lock down" and "only those qualified" route that you encourage, we'll soon see permission prompts to do anything anyway.

> That doesn't make sense. What is in / that is dangerous?

There have been examples of (un)installer scripts removing substantial parts of your system, steam for example. Linux in this case https://github.com/ValveSoftware/steam-for-linux/issues/3671, but not generally unthinkable on MacOS. It's generally good if the system puts up barriers to prevent such occurrences.

You're entirely welcome to disable these protections, but MacOS is designed to protect Joe User who may have substantially less knowledge and who has no general need to mess around in /etc

It seems to be solving a problem that doesn't exist, as if edge cases were the reason to apply blanket restrictions.

It's like banning knives because there's been a few stabbings. It removes the general utility due to misuse by a few.

"solving a problem that doesn't exist" ... um I nuked plenty linux machines by removing the wrong file early in my linux days. Hell I nuked a couple OS X machines when I first got Mac. Turning off SIP isn't that hard ... if you consider it hard you probably need to keep it on.

But typing the correct command also isn't that hard!

man synthetic.conf:

> synthetic.conf describes virtual symbolic links and empty directories to be created at the root mount point.

That doesn't actually create real directories; it's just symbolic links. For certain use cases (nix) this isn't enough.

Every new release of OS X has been an increasing disappointment for years now. Security issues, bugs, incompatibilities.

It's almost as if they've been actively devesting in OS X.

At least we now understand why.

Mac's are no longer machines fit for professional development. It's a damn shame, but it's true in my mind.

They just switched Macs to an entirely new architecture. I think they probably put more effort into macOS than iOS this year.

That's not my point. This new OS version is not OS X.

Neither was the last one!

The 10 in the version number is a clue[1]

[1] https://en.m.wikipedia.org/wiki/MacOS_Catalina


You got down voted because you didn't say anything. Go ahead read it again, I'll wait. Notice how the closest you got to making a point was Macs are not "fit for professional development"

I disagree and I think you're incorrect.

For a long time, I and other people I know chose MBP for professional web-development / video editing / music production because they were provided with a slick GUI on top of a solid operating system.

It's fairly apparent to most people who read tech news, that the quality of the operating system isn't what it was. For instance, think back to the 'no password necessary for root user' debacle a few years ago.

I don't think it's beyond the realms of credulity to consider why this type of problem was allowed to occur in Apple's 'professional' line.

If you're still with me, perhaps you'd also be able to follow my line of reasoning when I surmise that the reason for this degradation of support, may have been because they were planning on dropping their support for OS X .. with this convergence event being its replacement.

In reality, it's your own sassy comment that's devoid of content. Hope you really did wait! :)

Since you're using the 'abusive relationship' analogy I'd like to point out there are more actors to this beside Apple and you. Most app developers are honest, but there still are lots of 'bad apples' trying to infect your system.

Would you buy a house without any locks since it's convenient? As long as you have the key it's paranoid not to buy Apple because of some restrictions since any PC also has a lot of these (secure boot, DRM etc).

The house without locks is a bad analogy. A better one would be everyone is forced to buy the exact same house because government regulators and commercial lobbyists had decided what's best for us and an entire generation of brainwashed sycophants kiss the ground at their feet with each new decree they enforce.

> it's incredibly easy to bypass the stuff that makes a difference using the terminal

Not my experience. It took a couple weeks of noodling to get parts of my environment moved over, and some of my automation I just moved to a linux box because I got sick of it.

> we've somehow got to the point where people bash apple for actually taking privacy and security seriously


On the planet I use a Mac on, the complaints are about a set of non-manageable, opaque and somewhat buggy security changes that arbitrarily changed semantics on big chunks of the filesystem and broke a bunch of things is hard to figure out ways.

Add in that there is no sensible mechanism for cleaning up any of these magic permissions paths, and I don't even know what permissions are active on the machine I'm typing on.

I've loved OS X for being a decent gui on a real Unix. It had a decent run, but my next laptop will be a Linux box.

I'm in a similar situation. As a UNIX, command-line user I've noticed that every release they make things harder for us. Yeah, Homebrew is great, but Apple doesn't support it and they could break things anytime.

Lately, I've been doing most of my dev work on a Windows 10 machine using WSL. It's come a long way and is very usable now.

It’s unlikely that Apple break Homebrew. On the contrary, Apple is helping Homebrew tackle the architecture transition.

Having said that: I agree, WSL is pretty amazing.

> it's incredibly easy to bypass the stuff that makes a difference using the terminal.

Try running dtrace without having to configure a bunch of settings and restarting your Mac. Then you have to change the settings back and restart again to go back to the way it was.

It's a pain in the hole, and for basically no reason.

Isn't Instruments based on dtrace?

It is. dtrace outside of whatever Instruments needs to do its job is fairly crippled.

Easy to bypass was also an argument for locked down mobile OS. Software on these platforms is mostly useless aside from 2-3 social apps if you like that. I think locking down features is a mistake and does only draw in certain developers. The reason the pushback might not be as large is that competitors for desktop seem to be sleeping, or worse, try to do the same.

I still don't believe Apple on the privacy front. Yes, they are better than competitors. Anyway, I would like freedom and privacy and don't see them as contradictory. And you won't get more privacy if a manufacturer locks down environments.

We had OS without any privacy intrusion without any form of lock down. MS drops the ball constantly, agreed, but that isn't a metric anyone should look for.

People are drawing, creating music, making movies, editing photos on their mobile OSes. Saying that they are "mostly uses" does not make them so.

Freedom and privacy are at odds. If I give any app the freedom to access my messages or call list, does it increase or decrease privacy?

> If I give any app the freedom to access my messages or call list, does it increase or decrease privacy?

I have an app(lication) that has access to my messages. It doesn't do anything with them aside from showing them to me on request. It doesn't affect my privacy at all because doing so is far beyond its scope. If I had shitty apps, I would agree, but until now those seem mostly restricted to shitty app stores. But if they cannot be trusted, they cannot be trusted to see my messages at all.

> People are drawing, creating music, making movies, editing photos on their mobile OSes

True, they are not completely useless, just mostly. I can do that on other OS with far more freedom. Artists generally don't like restraints too much and I do actually paint digitally. Wouldn't know how to connect it to a mobile device, nevermind the atrocious software selection.

The cameras on mobile devices are nice though. Don't think the common mobile OS makes handling image date easy at all. Even most normal users want images on their desktop as soon as possible.


I agree completely. The only real gripe I feel is that excessive padding.

It is sort of like scrolling through Dribbble app mockups. The layouts look amazing as a mockup, but when you actually go to make or use the app it isn't as practical as you want.

On a desktop, with a mouse, I have near-perfect precision. I don't need overly-padded icons and windows because I have the precision of a mouse. The extra padding just wastes screen real-estate and makes me move the mouse more to get between tools or options in the menu.

I would much rather have less padding, which might not look as beautiful, and use the extra space for the content of my application as opposed to the toolbars.

> On a desktop, with a mouse, I have near-perfect precision. I don't need overly-padded icons and windows because I have the precision of a mouse.

You do, but not everyone does. Consider the accessibility needs of people with motor-degenerative diseases like Parkinson's or Huntington's. Won't they be better-served by larger click-targets?

You can always slow down the pointer movement itself to compensate for that loss in precision. With a well-chosen acceleration curve, that doesn't hinder you from covering even large, screen-wide distances with just a quick flick of the pointing device.

That is why Apple provides an accessibility panel in system settings.

Normally, yes; but sometimes the change required to help one group doesn't hinder anyone else, and also is very hard to automatically derive the "accessible" version of the UX in a way that doesn't cause more harm than help (by e.g. accidentally pushing critical buttons out of visibility into a slide menu; or breaking a menu onto two lines such that there's less viewport left than originally designed for at minimum height.)

In such cases, it makes a lot of sense to just give the change to everybody, so that you can design the UX once, with the change in place.

See also: modern TV shows and films that make sure to not put anything important where closed-captions would appear—they're designing once, assuming captioning.

If you want to make a UI more accessible to people that can't point easily the fix is a mouse with filtering and keyboard UI control, and crucially a way to click without moving the mouse. Not making the buttons slightly bigger.

doesn't hinder anyone else

Complete BS --- it absolutely slows down those who don't need elephantine UI elements to use a computer.

In this particular case, the new design is just trading element-internal padding for inter-element margin or element-group margin. The total space that each element-group takes up is the same. The only change is that more of the previously "dead space" around each element is now clickable, because it's now considered an internal part of the element.

It's a bit like making the UI into a https://en.wikipedia.org/wiki/Voronoi_diagram — any area on the UI chrome becomes "associated" with the nearest interactive element.

Like I said, I don't see how that hinders anyone.

I don't understand what you're inferring this from. To me it looks like the opposite is happening and dead space around elements is increasing (e.g. see the comparison of finder between 10.16 and 11.0).

Edit: It's also quite clear from the comparisons that the number of elements visible on-screen is decreasing because each element is getting larger and the padding between elements is increasing.

Edit2: Also, the idea that this UI is more accessible is contradicted by the removal of various visual elements that help separate and differentiate UI elements and the loss of contrast in the UI in general. Visual impairments are much more common than Parkinson's and none of these decisions help with that.

> they're designing once, assuming captioning.

At the cost of shrinking everyone's effective screen size.

No? They still put stuff down there. Just not critical stuff (e.g. establishing-shot location text, which instead tends to get displayed centered or center-left in modern visual design, since nobody ever sets their captions to the middle.)

Are we to believe that the UI is designed for people with disabilities or just hoping that this one aspect will make things slightly easier as an unintentional byproduct in between the time they roll it out and some future time when they change it?

Although 930k individuals in the US suffer from Parkinsons given the economic cost including medication, lose of income etc which averages about 24k per individual it seems likely that few of them are enjoying mac laptops even compared to the general population that is 10 times more likely to run windows.


Indeed this recent article is replete with references to cheap and budget not retina and style


> Although 930k individuals in the US suffer from Parkinsons given the economic cost including medication, lose of income etc which averages about 24k per individual it seems likely that few of them are enjoying mac laptops even compared to the general population that is 10 times more likely to run windows.

I would note that many people with disabilities use iOS devices. The layer-stack of iOS APIs has been top-to-bottom engineered with accessibility considerations in mind, rather than it being worked in after-the-fact; so accessibility features on iOS are a lot more “universal” than ones on other OSes. iPhones are basically “the” phone for the blind, for one thing.

Not saying that rebuts your other points. Just pointing out that when you have a disability, “cost” becomes just a matter of how long you need to save up for it; while “has the particular benefit” is the difference between the device being at all useful for you, vs. not. There’s no point in buying something that’s half the price if you can’t use it.

This is clearly not about accessibility but about trackpad users.

Just make it customizable. There is no one size fits all.

Unfortunately, Apple (and not just Apple, others are sadly doing it too) would rather change the users to fit its designs than the opposite.

This might be for apple to introduce touch screen on macs.

It's about time.

Might be good for laptops with a touchpad though. I don't know the sales split, but I assume Apple sells many more laptops than desktops.

I actually feel like I have even more precision on a touchpad because I'm not sliding around a weighted object

>Apple knows that a large percentage of the people who buy its computers are software engineers

I suspect a large percentage of software engineers buy their computers, but I'm skeptical at the assertion that a large percentage of people who buy their computers are software engineers.

Looks like they've been selling around 18 million macs a year (https://www.statista.com/statistics/263444/sales-of-apple-ma...). If you ignore individuals having multiple computers, and assume the average lifetime of a machine is 5 years, you get 90 million people with a Mac. I don't know how many developers there are, but saw 20-25 million in some estimates.

IIRC Tim said in today's keynote there are 23M Apple developers. Very roughly, 20% of macOS users are developers.

20% of users are Apple developers. There are plenty of developers who use their hardware but don't develop for Apple products (a web developer wouldn't be an Apple developer).

Or have things changed since I left the Apple ecosystem? I'm assuming you don't need to be a registered developer to use things like "make" or download IntelliJ IDEA?

Yes you're right, that's what I meant.

If you're doing web dev you don't need to be a registered Apple developer.

The vast majority of developed worldwide are using crummy Windows laptops and developing internal enterprise software :-) At best Macs probably have a marketshare of 30% among developers.

The US and big tech companies are outliers regarding Mac usage.

Win10 developer experience is far better than Mac OS. And way more versatile. Not to mention the brain dead design of UX when somebody FaceTimes you while you are developing in full screen mode...

At the last WWDC, Apple mentioned that of all of the so called "creative professionals" on the Mac, software developers were the largest group. I don't know how large the percentage is exactly, but its clearly significant.

Considering they intentionally got rid of, say, video editors as a professional group that used Macs, it may be more of a case of "who is left."

One of these days I'll write a todo list manager to replace OmniFocus and then be able to switch platforms.

I'm sure the only reason we buy Macs is so that we can still target iOS.

I honestly find MacOS to be a more usable environment than Windows. I haven't run Linux as my desktop for many years, though, so maybe I just need to go back and try it again.

That's probably true - although there are a LOT of iOS developers out there who can now essentially program for macOS because of catalyst.

Either way, software engineers build the applications that drive your platforms popularity, meaning developer adoption of a platform is incredibly important and thus they probably have a larger influence on Apples decision making than other demographics would with similar percentages.

There are more 'normies' than engineers out there.

They need to continue sell more to keep share holders happy.

It's a calculated shift.

Well, I'm usually the person to quickly point out that HN is very conservative, and advocate the modern UI/tech stack that HN thinks is too flat or too bloated, etc... But as a long time macOS user, I can't say that the UI is good.

If I can fix one thing from the new UI, I think it should have borders or at least a different background color, like pre-iOS 13 with button shapes. The current iOS buttons (and the new macOS design) doesn't give any visual cue for the user to know whether this is an icon label or a button. That's a problem, and one that macOS has solved elegantly by having different textures (like the toolbar).

It's now... more like Windows 10 where everything is on a flat surface without any differences.

macOS buttons do give you cues…when you hover on them :/

I’m finding the cohesive OS arguments a little ridiculous.

For one thing, it’s not like cohesion was achieved by bringing in the best parts of OSX and the best parts of iOS. Cohesion has been achieved by transforming the OSX UI into iOS’s to the extent possible. Apple has been clever to do it bit by bit but using OSX from 10 years ago to today, you can feel the alienation.

The other part is that until a few years ago Apple was the biggest proponent of the idea that a Mac, with its pixel perfect input mechanism, and a keyboard, and iOS whose input was fingers, should not have a similar UI. Their execs scoffed at Microsoft’s attempts to make a device which when in laptop mode ran traditional windows but in tablet mode ran a version of windows optimized for touch. Somehow those UI concerns have evaporated for no good reason.

Yep, I agree. Hacker News is a very conservative place (in terms of technology and UI changes). I think in general Apple does a phenomenal job at UI design.

That said, the user experience of Mac OS still has plenty of low hanging fruit:

- Window management and app switching could be much better. The "app switching system" I liked the most was in Gnome 2 / MATE - I could just quickly move the cursor to the bottom task bar and scroll. It was so effortless, fast and easy.

- When I download an app (.dmg) and click it, it shows up as a new disk. WTF?

- The file manager could be made much better. Again, MATE had (probably still has) the best file manager I've ever used in terms of UX.

> When I download an app (.dmg) and click it, it shows up as a new disk. WTF?

So download (or provide for download) a .pkg instead of a, you know, Apple Disk Image file

If I download a popular app and it shows up as a disk image, it's not really my fault. This is a UX fail. Apple has the power to fix this.

Normal people don't know and don't want to know what are apple disk images and what's the difference between .pkg and .dmg.

I think this is a case of "that's how it's always been done".

My understanding is that it's done this way so that there's a consistent experience installing software from a CD/DVD or from a download online.

But I think you're right, Apple should've rethought the software installation process the moment they dropped the optical drive from their computers.

I highly doubt "normal people" are losing much sleep over a virtual drive showing up in their Finder sidebars.

It's only a "UX fail" if you're of the opinion that all platforms should work exactly the same way, and/or that UX paradigms that aren't what you already know are inherently wrong.

I'm not saying people are losing sleep over it but it's just bad UX. If you download and install an app, it shouldn't have the side-effect of a new disk appearing on your desktop. It's obvious, ask any UX designer if you don't believe me.

Again, UX paradigms aren't inherently bad or wrong simply because they're not what _you_ are used to.

macOS users are aware (even though they might not know the technical terms in question) that a .dmg is not "an app". It's a virtual disk format - something reinforced by the literal disk that's its file icon - and using one is reminiscent of inserting a CD or USB stick to get files from. _That's the point_. You insert a disk and copy its contents to your own hard drive (applications aren't the only things stored or distributed via virtual disks).

It's like complaining that clicking on a ZIP file has the "side effect" of creating a new file/folder. People who actually use archives understand that they're container formats.

What percentage of Mac OS users know that .dmg is a virtual disk format, what's your guess?

I've been using Mac OS for years, so I'm used to it. And good UX works well even for people who're not used to it. See iOS, you just tap "install" in the App Store and the app installs, that's how it should be.

Seriously, ask a good UX designer what they think about this - or just ask normal people why they think there's a virtual disk drive on their desktop and whether they know what a virtual disk drive is. I don't have the time or energy to convince you.

It's not bad UX. Apple provides three other app installation options that don't create a new disk on your desktop. The fact the the developers don't want to use those 3 methods is not Apple's fault. Blame the developers for that.

Now if Apple where to follow you suggestion and lockdown the app installation for better UX, there would be pitchforks everywhere. In fact, the pitchforks where out just last week for this issue.

I blame both. Apple has the power to solve this in a way that doesn't have drawbacks for developers.

Yup, normal people just don’t care about it showing up as a virtual drive.

> When I download an app (.dmg) and click it, it shows up as a new disk.

How do you think it should show up?

As an installer program, like on Windows. Or as an installable package (Mac OS should have an integrated package management system like Homebrew).

DMGs aren’t installers, though, they’re disk images. There are separate installer packages, and disk images don’t have to contain software in them.

macOS installers exist; I think they were used more commonly before the .app format, but that's how you get the Windows situation where the "uninstall" operation is a crapshoot because relevant files are scattered across your system.

I don't know the exact purpose behind the "mount a virtual drive to install an app" model, but it seems far too deliberate to not have a purpose.

To be fair, there is no supported way to uninstall a .app with all of the scattered files, either (all of the things it places in ~/Library to run, like preferences).

For me, AppCleaner is one of the first things I download on a new machine.

For people who prefer the commandline: Homebrew can install not just open source software, but binaries as well, via Brew Cask. And often, you can choose to either uninstall an app with

  % brew uninstall <appname>
Or uninstall and remove all of its cruft as well via:

  % brew zap <appname>

I'm sure there's some rationale behind it. But it's simply a UX fail.

I barely even notice it now. I’ve been a Mac user since OS 8, and the workflow of mounting a disk image to install software has been around since before OS 8. Regular .pkg installers catch me off guard sometimes

I don't care what they do with the UI (unless they fixed it somehow.[1]) The reason I'm always disappointed to see a new version is the breakage that every MacOS update brings, both to developer tools and apps. I have to choose between keeping my various environments functioning and getting the latest MacOS so I can run Jira Desktop.

1) Delete spotlight, Siri and 'look up,' make alt-tab handle multiple windows of the same app, fix searching within folders and also separately fix whole-filesystem search.

Tip: Cmd + backtick cycles through windows within the current app. Once I learned this I much preferred the Mac's way of grouping windows by app.

I won't know until I actually use it, but based on looks alone I think it's the cleanest and most cohesive Mac UI since the peak Aqua era (OS X 10.4-10.6).

iOS and macOS should not look or act similarly, though. A UI should look the way it does in order to optimize efficiency and enjoyment. iOS has a touch UI for devices ~6 inches wide, and no keyboard. macOS has a KB/M UI for devices 13-34 inches wide with a keyboard and mouse, or worse even, a trackpad. You cannot make one more like the other without making it worse.

Levelling your UI to the lowest common denominator is a sure fire way of making it worse and reduces the efficiency and ability of everyone in the long run. macOS and iOS should be entirely distinct UI wise.

I agree with you in principle - but these UI updates were basically just refreshing the UX that already existed with a similar aesthetic look. They didn't disable the windowing system or make any other big changes in how the interface actually works. Just visual refreshes to bring them in sync. That's the thing I don't understand. Even the padding is a bit overstated by people, it's not that extreme.

My huge beef (yeah, I'm nitpicking, but it bugs me) is the camera icon.

The new one looks like a broken stop/play button (the play is backwards).

I know the old one was "skeuomorphic", but it was useful. This one burns brain calories just trying to figure out if it's a stop/play button or not.


>I really don't think they're going to put out touch screen Macs.

Why not? I think its a distinct possibility.

I think the "you can run iOS apps on your Mac now" announcement basically guaranteed touchscreen Macs. It might be possible to use many iOS apps with a trackpad, but it's not going to be a pleasant experience.

Put together a touch-friendly new OS UI with the ability to run a zillion apps designed-for-touch in a context where all your competition has touch-enabled devices and it's hard to believe they'd hold out any longer.

That's an interesting point, I didn't think about that. Plus, the extra padding would make more sense then. Maybe they are planning on adding it.

That's assuming that ios apps on the Mac are a big feature. They might just be some novelty to sell the move to arm, or a first step towards some sort of universal apps that just switch their ui according to the device they're used in.

Or it just provides a way for iOS app developer to quickly adapt their app for Macs. They mentioned introducing Mac UI paradigm alongside iPhone and iPad ones. So with some extra work you can target all.

Touchscreen Macs running iOS apps... At that point, what exactly is the difference between the product lines?

Would you like to go around with a laptop in your pocket?

If they were going to do this, they would have done it when they needed a big new feature and instead they put out the touch bar. Apple's answer to people who want something like this is the iPad Pro with the adaptive cursor and the keyboard case with the touch pad.

I could be wrong, it's just my instinct.

From the article: "Beginning in macOS 11, the track of a slider with tick marks aligns with the center of the thumb."

Ha, very good spot! Seems like another fairly strong hint (along with the padding and the iOS app support) that touch screen Macs are coming

Though someone on twitter pointed out that they believe “thumb” refers to the element that slides along the slider and it was always referred to as this, heh

> The extra padding in all of the UI is the only thing that I chafe at a little bit

This, and increased font sizes. Do any of you know how to change the font size of the window title bar in macOS 11?

It's interesting that this comment makes the same sentiment as the top comment on the CPU announcement thread - contrary to the (perceived) public opinion, things are good, it's going to be great. Can you both be really against the crowd if so many people agree with you?

(To be clear, I'm not insinuating anything, just interested in the dynamic.)

Agree - I think there's a silent majority thing going on here. Because I only saw negative comments on any thread I looked at about this. That's what prompted me to comment in the first place. Clearly, a lot of people do agree (or I'm just super persuasive :P)

People that disagree are usually the loudest when it comes to any issue. Those that agree probably just upvote and move on.

>It seems like people think that because they're taking design cues from iOS they intend to lock it down like iOS and they're transferring that fear into dislike of the new UI itself.

This has been the fear for the last 10 or so years, maybe longer. It hasn't strictly happened yet. Maybe someday, but this definitely isn't it.

It just wouldn't make a shred of sense. iPads have real keyboards now. A crippled macOS wouldn't have a purpose in today's world. It would make more sense to drop the category entirely than to dumb it down, and if they made anything clear today it was that the Mac is here to stay.

I liked it as soon as I saw the Accent Colors — feels like Appearance Manager in Mac OS 8 all over again :)

ive been using Mac OS X since 10.0 betas, I like every time they refine it. And this feels a step in the right direction. Simpler. I suspect everyone will get used to it and cringe at going back to the old one, just like always

as a UI geek these kinds of updates in Mac OS updates are some of my favorite - they impact everything you do in a small but substantial way

The loss of information density caused by the extra padding is a big functional change - a loss of functionality.

Even judging by that first screenshot, you can see that white text with very smooth shadow on a semi-transparent menubar is not at all pleasant or readable. For all the talk about "human interface" you'd think they would at least get the whole black-text-on-white-background principle.

Good points. Indeed this probably really helps with getting the 90% of iphone users who don’t own or use a mac yet to make it more familiar immediately.

More cohesive brand experience is not to be undervalued either both in terms of cost to produce assets and perception by users.

I like that the new UI changes. More white space and improvements to controls are welcome. I'm a bit partial to the lack of buttons that don't look clickable, but I'll reserve my judgement until I've tried it.

The problem is, Apple have chosen consumers above professionals.

It's been fairly obvious that this was their plan. Now it's been categorically confirmed.

To the downvoters .. hope that koolaid tastes good!

Virtually all of these changes reduce contrast and differentiation between controls while simultaneously decreasing information density. All bad news for people who work on their computers for a living.

I like borders on my buttons to know what is a button and what isn't, this feels.. backwards...

So much has to be inferred from the way things used to be, makes it harder to argue that it's the "simpler" operating system designed for Humans with a keen eye on UX design..

FWIW I also think the flat UI on iPhone (which has been prevelant for >6y now) is a horror show.

Steve Jobs famously once said (while working at NeXt): "There are two kinds of people at Apple: Those that want to push computing forward, and those that want to be the Sony of computers.. and the Sony guys are winning"

I remember the transition from iOS 6 to 7. It definitely initially looked modern but I still sometimes have difficulty distinguishing between a label and a button.

Another common issue is the lack of scroll bars or other UI hints like shadows means it's often impossible to know when/if you can scroll.

Though Apple isn’t relenting on this front in the default experience, you can change the appearance in Accessibility options (look for Button Shape). That could help keep these distinct and easier to recognize.

How on earth is moving to a the ARM architecture not pushing computing forward?

macOS moving to ARM is the biggest leap for developers in recent years - Amazon are also developing their own ARM chips and cloud services running linux ARM is the future.

Having a mac with a unix-like terminal environment & support to run linux arm natively when devloping for those services is a huge deal.

This is not the ARM thread, This is the UI thread. Nothing about the UI is predicated on ARM.

First of all, this off topic. Next, it's simply wrong, there is no ARM processor better than the best x86 processor in any category except insanely low power. All comparisons I've seen so far compare tiny mobile processors with huge server processes with orders of magnitude more I/O per watt. When you compare phone chips with Zen 2 low-power laptop chips and take into account the massive I/O disparity it's much less favourable.

I thought the “white everything” UI approach of metro apps under Windows 10 was bad enough but this is even worse with all the bluriness and similar colours. I am going to have to get another job soon else I’m going to go blind with all these “improvements”. I’m not joking.

i've had "increase contrast" (and implicitly "reduce transparency") enabled since whenever I started using macOS. I really hope those accessibility options work well to reduce whatever changes they're trying to bring about.

on the same note; the"decreasing information density" I don't think would be a problem if you're switching between the "terminal/IDE/browser/excel". the "table views" still seem fairly "compact" so..? /shrug

They can simply get rid of the close/minimize/maximize buttons at this point. I wonder just how many people actually ever use them.

At this point they’re just about the only thing that tells you whether a window is in the foreground or not.

For the last few versions of iOS, MacOS has felt somewhat old-fashioned, lacking things I've grown to really appreciate on my phone and iPad like the Control Center. I'm glad that is now coming to MacOS. This feels like a much-needed facelift.

I'm glad is isn't a polished XFCE, there are a lot of options for people who like this sort of thing already, just run Linux. In its current state, Linux on the desktop is definitely not for me, but then, I like the touchbar (as a software engineer, even), I like their new keyboard better than the pre-USB-C one (the butterfly one was terrible though), I'm fine with four TB3/USB-C ports, I don't miss MagSafe much. I like not having to bother with a myriad settings, because the defaults are usually just fine and most things can't be fine-tuned anyway.

Big Sur seems to be a really interesting spin on a desktop iOS that embraces and extends the security and UX concepts from iOS, but for keyboard/mouse and without being so locked-down and constrained. From what they've shown so far, I like it.

Now please have iPadOS become more like MacOS – that would be terriffic.

I expect that MacOS 11.2 will be iPadOS.

strong disagree. MacOS and iPadOS will take cues from each other, but they will remain fundamentally different. MacOS for the old-timey folks who want root access and to be able to browse the file system, iPadOS for people who like having all for that abstracted away from them a-la iOS. neither is right or wrong, just caters to different kinds of people who prefer different ways of working.

IMO you vastly overestimate the care Apple has for "old-timey folks who want root access." They will remove your usecase in due time, when they decide to merge the two(MacOS and iPadOS).

They could have done that at any point in the last 10 years. They're not going to. "Operating systems" are the bit that makes the hardware work for the software, and because the hardware of "computers" is different from "portables", they'll continue to get different OS for a LONG time. Apple make mac-, i-, iPad-, tv- and watch- OS. That's five different OSs. The "bringing together" of macOS and iOS has nothing to do with unifying the OS over the hardware: it's about unifying the experience of interacting with digital representations of physical reality. Need a surface for displaying 2d media? Here's a "window" or an "app". Need to write a message to your friend? Here's a surface that accepts typed characters or your handwriting. Need to write some code? Here's a surface that accepts text and has some editing controls. Etc.

What's much more likely is that they'll find some new form factor of sensors and projectors (keys, screens, pointers, loudspeakers, data i/o) that does away with "computers" and "tablets" entirely. And that'll have its own widgetOS.

EDIT: punctuation.

Maybe they will, maybe they won't. Right now they don't appear to be moving into that direction – what better opportunity to further lock down things than a big architecture change, when you have to rebuild lots and lots of things anyway, when you're transplanting a whole bunch of iOS features (and even whole iOS apps, it seems) to MacOS?

But instead they explicitly mentioned virtualization support for Linux. There seem to be no additional measures like Gatekeeper and Notarization. The security measures introduced in Catalina have carefully-placed gaps to not break terminal workflows, and unless I'm mistaken they can be disabled. Notarization is a bit of a mixed bag; I get the security benefits, but it does give them quite a lot of control.

I may be totally naive and utterly wrong, but the vibes I get from all of this isn't that Apple is currently pursuing a strategy of forcing MacOS into the iOS mold and turning it into a two-apps-side-by-side, fully locked down, anti-productivity, content consumption and light Excel, iPad-with-permanently-attached-keyboard device.

Rather, they seem so be trying to build a nice and coherent user experience across phone, tablet, laptop/desktop on the UX front that works for a wide audience with little customization, and to find a security compromise between hundreds of millions of (potential) users and organizations who need an ecosystem that keeps them as safe as possible vs. developers, researchers etc. who want their systems to be very open and unrestricted and safeguards to be very lax. I imagine it is kinda hard to build a product that serves both of these well while remaining coherent and fun to use.

With that in mind, I think they did a decent job at this with Catalina and the lastest iOS. Big Sur seems to be a continuation of that; it remains to be seen what it'll actually be like in the beta, but right now, I'm optimistic.

Someone has to write the OS…

There's also the fact that macOS's UI isn't touch-based, which is a distinction that apple follows religiously. As long as they keep that distinction, there will always be a need for both UIs.

Judging by all the whitespace in the new guidelines, they are in fact preparing for a touch-based macOS.

Larger rounded corners, taller toolbars, and more padding within sections all further reduce the amount of actual content you can display. If you look at most of the before/after screenshots, there's clearly less information delivered within the same amount of screen space. That's nice for a touch interface perhaps, but it's counterproductive for a computer.

And making toolbar buttons all monochrome outlines does make toolbars have a more consistent appearance, but that actually makes finding the button you need more difficult. The differences are what we notice, and later what our brains automatically key on when we need to actually use the interface.

Less than half of these changes are arguably worth having. The rest seem like marketing saying, "We need to be fresher, guys!"

Using a new interface can be fun because of the change, but that fun quickly wears off. Then we're left with the realities of the interface. That Finder window is going to be much less useful in the real world with real file/folder names that are longer than 12 characters wide.

This is a UI for casual users. It might be good for grandma, but it's less good for people who need to get work done.

Two observations regarding the former title bar:

* Putting the title (folder name) before the controls introduces a fundamental instability, much like OS X did with the menu bar by placing the named app menu before any other menus. Also, the amount of available space for displaying a longer title/file name seems limited.

* It seems, the proxy icon is gone. I can't see which element could bear the functionality in place. The title is already editable, making it dragable as well might be too much. Is this the beginning of the end of the drag-and-drop paradigm? (Notably, it doesn't suit touch interfaces well. Many aspects of these changes seem to prepare "touchability".)

It was featured in some article on HN before that we used to have "safe areas". Areas where you could grab a window and drag it around, without having to carefully place the cursor or accidentally pressing a button. This is not new in Big Sur, but has been happening for some releases now. Take the Notes.app, it doesn't even have a title bar, just a tool bar. When resizing it to (what I think is) a comfortable size, the toolbar buttons are side-to-side leaving just a few pixels of margin by which you can drag the window. I'm not a fan of this trend, but it seems the way forward for all macOS apps. With this release it is brought to most first-party apps.

The Proxy icon is a seriously underrated feature on OS X.

Wish every other OS would replicate it already.

Apple made a pretty big to-do about adding drag and drop to iPadOS. I don’t see drag and drop going anywhere.

The proxy icon is still there, it seems you have to hover or click to show it though.

So discoverability is finally replaced by the "Ahhh" paradigm?

That's been the trend for a while, but in the case of the proxy icon, the types of interactions you could perform with it were mostly undiscoverable and unknown to a lot of users. You had to actively explore the way the UI worked to figure it out, but I can't say I'm a fan of the change either.

> in the case of the proxy icon, the types of interactions you could perform with it were mostly undiscoverable and unknown to a lot of users.

That's a good point I hadn't thought of -- I love the proxy icon. I'm glad it's still there, but this kind of feels like Apple is implicitly saying, "Look, for the ten of you who knew what this was, you still do it, but we're just hiding it from everyone else, sorry."

It’s easy to take every UI change as an affront, particularly when somebody changes something we liked. Just from what I’ve seen though, it appears the goal is a simply a more space efficient UI. It’s not a bad goal, even though I still want to beat them over the head with the Dock which functioned better under NeXTSTEP than it ever did under Mac OS X, and the stoplight isn’t doing anyone any favors over the Mac OS 9 window controls.

I guess I shouldn’t be fighting the wars of 20 years ago though. It is what it is, and since the goal seems to be to rebuild everything in SwiftUI eventually, maybe the Dock will finally be reconsidered in the next few versions when they get around to it.

The problem is, it seems to be not so much about space efficiency, but about streamlined appearance (e.g., there's now less horizontal space and fewer controls are shown at once), which translates to low semantic contrast, wich again translates to lower work efficiency. And it's the latter, which causes most of the criticism, as not all users want to trade personal time and usability for looks (or prettier printouts of the UI).

I'm not going to defend Apple's work on this front, it's just they decided to consolidate the title bar and toolbar to free up some vertical screen real estate pressure and they're leaning hard on these slide animations to hide some of the UI elements so as to not overload the toolbar/titlebar thing. That to me looks like a botched attempt to make it more space efficient, although I do at least like the bit with the search bar.

John Siracusa has a different theory though, and having had time to go through and examine to changes in the HIG in more detail, I'm inclined to agree. The spacing and layout and sizing appears to make Mac OS XI more "touchable", i.e. we should expect Macs with touch screens at some point.

I can also see, why this is not a bad choice for touch interfaces.

However, at the same time, this also illustrates, why it may be not that great a choice for a WIMP interface: the kind of "intimacy" related to the specific mode of interaction is essentially different. Where contrast and too much of variety may be irritating for a touch interface, it may be already too low in contrast and information density for a WIMP GUI. On the other hand, visual grouping (and semantic collapsing of groups) is much more important for the more remote WIMP interface, whereas touch already is much more locally focused. Generally, treating them both the same isn't a good idea.

I don’t mind the stoplight. I do mind Apple repurposing the god damned zoom button for fullscreen, requiring I hold option every time or having to install third party tools for what should be at worst a freaking plist change.

A circle given the same height provides a smaller pointer target than a square. The stoplight was also a downgrade over the shapes, positioning and functionality of the Mac OS 9 window controls. Stickies still uses these controls by the way, if you want to see what it is like. Close is all the way over on the left and its a box, I would prefer an X ("batsu"), but it's a box. WindowShade-style minimize is all the way over on the right, and Zoom, well it's the triangle but its been degraded in functionality. At this point it's maximize and revert, provided you don't resize it from its maximized state.

Re: Zoom. Maybe you already know this, but just in case you don't, double click the titlebar. Even works in Safari, although there's less "titlebar" to work with. If you already know all that, well, maybe it will help someone else reading through the thread.

That said I did use BetterTouchTool to modify its behavior because I too was frustrated with changing functional behavior and I recently bought a three button mouse to use in addition to my other input methods, but I decided to make it default Zoom, but middle click Full Screen.

You can also set a keyboard shortcut if you like in the Keyboard settings (no third party software required) within System Preferences. Just go to App Shortcuts, All Applications, add a shortcut, for the Menu Title type Zoom and then just record your shortcut.

>A circle given the same height provides a smaller pointer target than a square.

To be fair, the click target of the traffic light circles isn't limited to the actual drawn pixels. You can click a fair few pixels outside of it and still get the same result (basically wherever the hover state activates).

Yes but given the same margin for error, a square would still be a bigger target.

The stoplight was a pure aesthetic choice, I would have preferred more functional choices, but it was probably the right call at the time to go for aesthetics. A new aesthetic on a new operating system, that’s exciting, that gets people talking. That said, it’s one of the last elements of Aqua that has remained almost entirely untouched, through pinstripes, brushed metal, stitched leather and now apparently, iOS 7 Strikes Back or whatever we’re going to call it in Mac OS XI.

But most macOS users don't know about the proxy icon either. It was never "discoverable", in the way you suggest, to begin with. The Mac has had "ahhh" paradigm features for YEARS. You could even say that the "wow" of the genie effect and the "ooh" of Aqua were cousins to the "ahhh" paradigm.

I love the proxy icon, too, by the way. But it's always been a badly documented, surprising, quirky feature. I found it by accident, by slipping and dragging a folder onto the menubar. That's not even useful; I didn't understand what I'd done. Eventually, after reading blogs and manuals, I use the proxy icon (and the brilliant Finder<->save dialog tricks) all the time.

I don't think we gain much by framing the argument around "discoverability" any more. I think it's much more useful to talk in terms of the overall interface paradigm. The proxy icon kind fitted into the "desktop" paradigm - i.e. a scatter of media and tools strewn across a desk. But Apple are moving away from that, and into a new paradigm of "surfaces for interaction". The i-devices have been in that space for years. The "mac" devices are now going there. Both are gonna change again, bit by bit, in the future as humans develop new ways to interact with this weird networked functional data thing they've created, and Apple try to provide novel tooling for novel technology. They do seem to be going faster, and breaking more things, than they did in the 80s and early 90s when they lead the desktop paradigm. They're possibly experimenting with ideas on the wider user base, rather than doing closed door research. A big part of that is due to the wider user base being MUCH more comfortable with computing than before. People with very little "computing" training are contributing daily to the development of digital culture.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact