Just one anecdote, but some reports definitely get looked at.

I assume this would have landed them in hot waters with EU privacy regulators, so they were very keen to replicate the bug and then have me check if it no longer happened.

On the other hand, the "first screenshot fails to display screenshot preview and doesn't flash the screen" has been in iOS for 3 versions now. I've reported it thrice, and no one has given the tickets a second look, marked it DUPLICATE, or anything. And every time I mention it, at least 3-4 other people comment on also experiencing the bug, so I assume Apple must be pretty aware.

My suspicion is that there's a really gnarly race condition as the root cause, and they haven't been able to find it.

You will never get a response to tickets unless it is affecting a lot of people and they need more information e.g. crash reports.

They are all read/triaged though.

Granted that was over a decade ago, and "no QA team" doesn't mean no testing, but given the numerous bugs in macOS today, and that they almost never get fixed, I'm not surprised.

(FWIW, I do not experience this bug you mentioned)

To be fair, Apple can still pull off the occasional amazing feat of vertical integration -- HDR, APFS, keeping audio latency under control despite the relentless assault of apathy from all directions -- but they never had the same level of consistent drive forward, at least not until a year or two ago when the big push for AI integration started. Apple gets ragged on here, but I think their integration is actually some of the best. They were putting neural cores in chips back when that sort of thing got mocked, not lauded, and every step has been thoughtfully tied in rather than airdropped from a ChatGPT science fair project. But they never got good at building or deploying leading-edge models themselves; I hope they turn it around because this is important.

See my longer comment here: https://news.ycombinator.com/item?id=43249634

You felt a real sense of ownership to the thing that you worked on. You worked hard and fixed bugs because it felt like it mattered, because you thought about how e.g. your mom would end up using the product, and also Steve Jobs would see it, so it had to be great. Also, teams were small. Something would involve only 1-2 people, and then we would look over at Redmond and they'd have dozens of people working on the same thing. The need-to-know secrecy was not just for PR value; it helped keep circles of communication tight, cutting out a lot of noise, so you could just focus. The organization was stable (and relatively flat, around 5 levels from junior engineer to SJ). I think in my 9 years or so there, there were no major reorgs. Avie phased himself out and retired, and Bertrand moved up. The only major disruption was when the iPhone project happened.

Release cycles were annual. Throughout most of the release cycle, it was pretty free up to each team and engineer to decide what to work on and how to prioritize it. Near the end of a release, it would get more and more strict on what you were allowed to change, up to the point where Bertrand sometimes would even ask to see code diffs.

I don't really know what is going on over there now. They have moved to a more agile approach, with more frequent integration checkpoints. In theory this should be better, but I suspect there's less sense of ownership and more of a feeling of a software factory. But it's probably mostly to do with the fact that the systems are way more complex, both the tech and the org, with way more moving parts. Even the programming language itself (Swift) is a moving target. I know (from talking to friends) there's a lot more politics and career-building going on, the kind of corporate douchebaggery that would not have been tolerated under Steve Jobs. People are thinking about RSUs and their promotions, rather than the products.

Ultimately, I think it boils down to this observation by jwz at Netscape, that there's "two kinds of people: those who want to go work for a company to make it successful, and those who want to go work for a successful company." Post-iPhone, Apple has filled up with the latter. A majority of the people at Apple now didn't work there under SJ, and the senior management who did experience that is now aging and retiring. At least from the outside, as a customer and end-user, it feels obvious that the founder-led product-obsessed culture is gone.

Open Activity Monitor, and type in System Settings into the search. Then open the Settings app and press the down arrow key through all of the menus. You'll notice that each one of them appears as their own line item in Activity Monitor until you quit Settings, and if you keep going up and down through the menus, it'll (probably) get slower and slower; it seems like there's a memory leak or something going on there, and my hunch is that each old settings menu was thinly wrapped in a SwiftUI view and gets launched as soon as you click its nav item.

So is Swift just too crappy for this kind of UI, that accesses low level system stuff? or are the devs just incompetent? Who knows…

I think Apple has decent QA for low level stuff (remember how they quietly converted every iphone to APFS and back just to test that it will work later) but bad QA for final GUI.

Most of cocoa/objc stuff was written long ago and back when QA was better. If this was cocoa and objective c today it would be equally buggy.

True, imo.

> But even that doesn't matter > Most of cocoa/objc stuff was written long ago and back when QA was better. If this was cocoa and objective c today it would be equally buggy.

I don't think I agree about this entirely. Although you can write unperformant code in any language and framework, SwiftUI is a different world of heavy abstraction. Writing a list view in AppKit using the documentation available allows for and encourages much more deliberate implementation in such a way that it'll be an almost night and day difference compared to the likely NavigationSplitView they're using in Settings. It's an inverse relationship with performance that SwiftUI takes; efficiency of piecing together the high level details first, and little to no control or clarity on how you'd optimize for performance.

My hunch is that with Settings, they're using a NavigationView or NavigationSplitView and not letting each view be recreated on selection, so they're sticking around in memory. However, if they were recreating each pane on selection, it seems like the render of the view waits for some kind of I/O, maybe reading their .plist or interacting directly with a system level CLI, which is why there's a delay before the UI renders.

Another possibility is a combo of both, they're retaining the views in memory AND triggering I/O immediately on selection, so the reads and threads accumulate quickly without being released

At least some of them are actually web pages:

https://blog.jim-nielsen.com/2022/inspecting-web-views-in-ma...

I'm not saying this is what happens. But it's scarily plausible and it really shouldn't be.

And yes, as you browse around system settings there are TCP connections to MS servers. I don't know that they are "blocking" though and have run out of care factor.

I’m sure the new UI piled on top doesn’t help, but as a consolation, the core software is shit too.

Chrome removed ublock origin for me today and I thought to myself why am I even on this OS anymore? What’s keeping me here? Decided to use the outrage over that to just make a clean break from Windows too.

I installed Ubuntu tonight, but this time I’m sticking with it. It already feels so good to have the software behave in a way that makes sense and isn’t some dark pattern meant to harm me and extract value somehow from me.

[1] https://getaurora.dev/

All non-core software is provided by Flathub (through the Discover app) and Brew (for cli apps). Both have worked flawlessly for me as well.

Ubuntu pushing snaps like MS was the last straw.

Dude... Mint, Arch / Manjaro, Void, even Alpine is better. Plenty of good and very lean+mean options out there.

Ubuntu is the Windows of the Linux world. Has been for several years now.

Try looking up what Canonical did and still does with the Snap store.

An example: new PowerToys https://github.com/microsoft/PowerToys/

The FancyZones "window snap" UI takes upwards of half a second to activate when dragging a window and the Zone Editor is at around 5s. All in all it is only very slightly less buggy than 3rd party tiling WMs like komorebic.

The PowerToysRun utility input is extremely variable, takes between 1 and 20(!) seconds. A lot of the plugins shipped with it simply don't work or have no suggestions/hints once you enter their prefix. The search relies on WindowsSearch, which is about 500x slower than https://github.com/sharkdp/fd and has not improved since Win7. Who cares, nobody ever searches for files, right? As a whole, PTRun is simply worse than https://github.com/Flow-Launcher/Flow.Launcher which uses the same UI kit as far as I can see. WTF?

There's another app that does this in the registry I think, but I keep forgetting the name of it.

https://learn.microsoft.com/en-us/sysinternals/downloads/ctr...

It's the first thing I download whenever I'm setting up a new Windows machine.

Go have a tour on Github for the endless collection of issues, some of them 5 years old, when Project Reunion was announced.

Unfortunately, to stay on top Apple doesn't have to do well for the customer. They only have to do better than wintel/android machines.

Control panel has sucked since windows 2.0

So far, I am not willing to ship anything in SwiftUI. I don't think it's up to the task.

I know that SwiftUI has some native wrappers, like maps, but the way that SwiftUI works, is so radically different from UIKit, that I think mixing them is problematic.

also - it's such a bummer that they have decided to shit the bed so hard on software at a moment when their hardware lineup is arguably at its pinnacle. like, the hardware has been firing on all cylinders since M1 but the software degradation is making it less and less pleasant to use.

And god, don't even get me started on how the icons rearrange themselves when you're organizing your home screen / control center. I can't believe they actually shipped it like that and still haven't made it any better.

If you try moving one of the icons anywhere, it snaps back to the top right corner right away.

I haven't found a fix. The only fix is moving the icons away from the Desktop into another folder using a Finder window.

Tried dragging color cursor for 30 seconds+, no issues at all. MacOS 15.3.1 (24D70) on 16" M2 Max.

Sequoia 15.3.1 (24D70)

I see this Custom Colour thingie at the top-right corner of the Wallpaper section, above a "Show on all Spaces" checkbox and left of a fairly big representation of the current desktop background.

After a bit of tinkering: this Custom Colour element is replaced with something else depending on the current background mode. If you selected a wallpaper image, it shows the name of the wallpaper. If you select a predefined colour, it shows the name of the color. When adding a custom colour, it will show an interactive element which allows to change the color in place, and that shows the buggy behaviour for me.

Ok, this at least explains why other people don't stumble over this as an obvious bug, I assumed it would be obvious, because the first thing I always do on a new Mac is to customize the background color by right-clicking the desktop, and since that moment I have that buggy Custom Colour element sitting there.

Not a great UX either way though.

PS: ...and now after adding a new custom color via the to bottom row of predefined colors, the bug in the 'Custom Colour' widget is gone and nobody will believe me it was ever there. Great :D

PPS: nope, it's coming back after going through the 'desktop => right-click => change wallpaper...' route again, phew.

In a way it's a perfect little example of how a bug can seem obscure to some (most) users/developers but seem glaring and unacceptable to a few (the few who happen to use the relevant feature a lot).

[1]: http://euclid.psych.yorku.ca/SCS/Gallery/colorpick.html

I’ll grant I might just be mole-eyesing it, it’s been a long day.

It seems the difference is that with the "custom color" button, Settings applies the colour directly to the background, whereas the plus button at the bottom only applies it when you're done. Applying it directly seems to be computationally expensive (ass various elements of the UI need to figure out whether to render their text in black or white, depending on the colour - would be my guess at least).

I mean for god’s sake these morons (yes, “morons”) have not yet figured out how yo sync browser tabs which is something new browsers get right in a few days to few weeks time, and sometimes on top of their incompetently done iCloud and related SDKs.

Apple sometimes comes across as a glasshouse built as marketing, too much money, (sadly) a huge army of fans and loyalist apologists (and not demanding customers), and an absolute lack of decent competition; and the biggest of it — a deliberate attitude of non-openness!

I mean everything Apple is closed! So how can anyone even quantify how bad their iOS is, how smelly their cloud suites are, how ridiculous their security is!! If you can’t see what happens behind a wall and the entity behind that wall has money more than most nations and a PR and tech propaganda machinery rivaling some of “those” nation states, how can you even be sure!

I feel like a lot of the Apple issues come from the fact they keep building on top of iCloud. It’s only very recently that people started trusting the sync.

It’s like MS, where anything built on top of sharepoint is going to be garbage.

BTW I have found a workaround using BetterDisplay and an EDID override (to more closely match what the monitor is actually telling macOS).

Related bug: macOS defaults to variable refresh rate when available instead of remembering my choice of 144hz. This is confounded by my hub (Caldigit TS3 Plus), which has trouble with variable refresh rates that result in a black screen.

The cherry on top: either I use a HDMI cable and deal with BetterDisplay forcing RGB to fix YCbCr, or a black screen when using DP through my hub due to the above bug.

Sometimes I wish Apple would get broken up just so macOS could have a chance at getting more love.

https://www.anker.com/products/a8317?variant=42329259475094&...

Does your adapter work at 120Hz without updating the firmware? If it does, does it support HDCP?

Yeah I had to flash the firmware on the Anker dongle, following this guide: https://forums.macrumors.com/threads/dp-usb-c-thunderbolt-to...

I just tried streaming 4K through the Apple TV app on my M1 Macbook Pro and no issues, so I'm assuming HDCP works.

As much as I like the integration between the phone and macOS I like the idea of desktop Mac getting more love.

Example: iCloud Photos syncing is complete crap on macOS. If it has synced recently it’s not going to do it again. So you sit there like an idiot waiting 10 minutes for a photo you just took on your phone e to show up. When a pull to refresh or refresh button would have fixed it.

FWIW, here's the hacked script[0] which only keeps the EDID data patching part. Be warned it's very hacky with the base64 EDID to be patched hard-coded in line 8 of the script. It prints out the patched EDID base64 which should be entered back into BetterDisplay (which is also where you can get the unpatched base64 EDID).

[0] https://gist.github.com/karmakaze/f795171a6a795491e754c3d092...

Either way macOS has no trouble with my 27” 2560x1440 Asus and Alienware monitors. Both connect with 10bit RGB no problem, at least over USB-C and DisplayPort (haven’t tried HDMI).

2560x1440 is a strong indicator of a monitor, but 4k over HDMI tends to get detected as a TV

I cannot get osx to actually output at 1080p, all it does is output at 4k and scale the result.

The downscaling in the projector adds input lag and just drives me crazy. I really wish they'd just let you control these things rather than poorly guessing.

I didn't know about better display, I guess I should try it and see if it can fix this problem.

Also extra fun is guaranteed if one end of the video cable is encoding with e.g. BT.601 primaries, while the other end is decoding as e.g. BT.709, or vice versa.

[1] https://www.microsoft.com/en-us/research/wp-content/uploads/...

It was the poor non-crisp pixels that led me to even knowing there were even RGB/YPbPr modes.

BetterDisplay provided a workaround, but it needs to be selected every time the monitor is hooked up.

(I guess that's normal for Apple stuff nowadays - when I hook up my ipad to my projector, I need to tell it every single time not to use the audio output of the projector, but keep using the bluetooth speaker.)

Look at the state of Xcode, a tool that's fundamental to the iPhone's appeal. Every developer knows that this thing needs a massive rewrite. The word is that nobody within Apple even understands it thoroughly anymore, so it's way past time to strap it on and build a modern tool from the ground up that's maintainable, instead of slapping band-aids on Project Builder indefinitely. Come on, Apple, you can afford to throw resources at this for a year and just get it DONE.

Meanwhile, Apple is letting open, "urgent" QA personnel requisitions sit unfilled for YEARS. We can all see the results.

Data-point of one, but: the new Settings panel is very much appreciated by the non-technical users I support, because it approximates the iPhone settings panel they're accustomed to. Personally, I hate it, but I also like that my colleagues are better able to discover and make changes for themselves. I figure learning new tech is a core part of my job description (but not theirs), so my frustration counts for less than their comfort, so I suck it up.

Xcode, however? I completely agree with you there.

https://en.wikipedia.org/wiki/Reality_distortion_field

now reopen the wallpaper setting again, click on top right custom color and do this, somehow the behavior is different. Now it change the wallpaper color as you drag over the colors rather than mouseup. My guess is clicking on + button at the bottom is triggering the popup config to update on mouseup, while opening it fresh will configure it to trigger on update, until the + button is clicked.

Reminds me of the butterfly keyboard issue of the 2016-2019” model year MBs.

If your warranty is still active could try to get the trackpad replaced.

I honestly have some work to do this month.

"Hey, let's rewrite the framework again and not do any regression testing or test against old implementations or see what happens with any codec that is not exactly what we are expecting for any reason." - Airplay devs, every year.

Edit: Repro-ed using the additional steps you mentioned below. As someone who handles external bug reports and writes them, it's so often the case that there are additional steps or a specific start state required, which both prevents reproducing the bug and narrows the affected user base.

rawrrrrrrr

Yes, but code reviewers signed off on it, our unit tests have 100% code coverage and they all passed. It must be okay.

https://gally.net/temp/20250304macoswindowopeningposition.jp...

I understand your point, once you have absorbed the Apple logic it is reinforced continually and makes sense. But opening a dialog near the use interaction point is a very reasonable expectation.

For me, I’ve never noticed any logic to it, I just know I need to hunt for it.

* in safari private mode, open image picker

* switch to different app (e.g. go to WhatsApp to save a new image)

* go back to safari

the image picker can now no longer be spawned from that safari private tab, you'll have to open a new tab to re enable the image picker.

EDIT: can repro. But it's very important to note: you need to have a color selected already. So select a color with the + sign, close the app, open it, top right click custom color, and then the bug appears, although not like you describe: the color selection is easy, it jumps around for 20ms during dragging. If this is what we call "low quality" I'm happy to stay on Mac.

Might be a reason why it wasn't fixed if you didn't include that vital step in your repro.

After initial custom color selection by clicking on the "+" (which works fine) and then reopening the Window to click "Custom Color" and then selecting the color again... it doesn't just jump for 20ms - it goes into full on psychotic flash jump behavior and basically continues to do so hands-off for 5 entire seconds before it stops at a random color.

There is 2 options now:

1) nobody has ever tested this workflow at Apple, automated or not.

2) it was tested and discovered but then pushed into the backlog as non-priority. Here the question arises - for how long?

The bug probably generates zero lost dollars so nobody at Apple cares anymore. THIS is what used to be different.

The other big time sucks are C++ interop and dozens of micro-dependencies (react-native).