Live data from Hacker News

The End of OS X

stratechery.com

51–60 of 589 posts

Re: The End of OS X

#51

I know this is about the coming MacOS 11.0, but Catalina was the end of OS X for me, and the end of Macs in general for me. I will not be migrating from Mojave to Catalina, and will use my current generation of Macs until they are obsolete, while finally making the transition to Linux. Mac and I had a good run, but I'm clearly no longer the target demographic and their corporate goals aren't aligned with mine as a cu…

Same here. Still in 10.14. Work machine is already on Linux. Personal one will be soon. Sad to lose some apps (e.g. Dash, Affinity, Alfred, Fantastical, Postbox, 1Password, ...) but in the end it will be more powerful and fast with Linux and i3 as I spend a lot of time inside the terminal and text editors. My 2014 MBP is slowly dying from hardware problems (screen goes blank at random, keyboard wears out, cursor jumps at random from time to time, ...). I am bored. Made a detour with Apple for a few years and it was fun (beautiful and Unix based) but I'm coming back to Linux with a minimal setup. When it comes to servers, I've never left Linux. And last but not least I prefer free software, open source, the mentality around it than closed proprietary stuff. I have a dying iPhone too.

Re: The End of OS X

#52

> what the company perceives as a weakness: the paucity of apps in the Mac App Store. I think that the issue with the Mac App store, is that it imposes restrictions on a once-free-for-all development environment. I know many app developers that have abandoned it, after giving it a damn fair shake. Mac developers just got used to being able to wander all over the house, poking their noses into whatever they liked. iOS…

On the other hand, what a modern desktop is sorely lacking right now is protecting users and developers from themselves. Whenever I run "npm install" I have to close my eyes and ignore the chills down my spine thinking about how a thousand random npmjs.org account holders just ran their postinstall scripts on my computer with full access to ~/Documents/. The traditional multi-user Unix permission system is no longer appropriate for the modern waterhole malware world. Nobody cares if you break root and hack system binaries, they can always be restored. The gold is in the user's content. We need a system where regular users don't need to worry about defending against software developers who don't carry the user's best interest in mind, and we need a system where even the cowboy I-know-what-I'm-doing developer doesn't point a gun at their foot, while still providing the freedom to tinker and share.

Re: The End of OS X

#53
What I think a lot of us programmers miss is the experience of the end-users who are likely to buy a Mac.

The fact of the matter is, 99.9% of people who buy a Mac are not going to know what subpixel hinting is. Or care too much that they can't delete Chess.app. To them, the Mac is nicer looking and easier to use than ever.

Will iOS devs care? Not at all - developing for the Mac has become easier with Catalyst. It's easy and cheap to get a new audience for your apps. Will web devs care? Not while homebrew and all of the packages in it still work fine. I, as a web dev, have everything working great on the Big Sur beta already.

And as for some saying that MacOS is becoming more iOS like: Yes, it's been happening since Mac OS X Lion with the fullscreen Launchpad. Nothing has changed.

Re: The End of OS X

#54
post #8

I know this is about the coming MacOS 11.0, but Catalina was the end of OS X for me, and the end of Macs in general for me. I will not be migrating from Mojave to Catalina, and will use my current generation of Macs until they are obsolete, while finally making the transition to Linux. Mac and I had a good run, but I'm clearly no longer the target demographic and their corporate goals aren't aligned with mine as a cu…

I suspect you're being downvoted because this is an empty, meaningless comment without some elaboration on why you've decided this. As it stands, there's no room for discussion: the only paths forward are for others to simply agree or disagree, or to try their best to guess at what your concerns are in order to further the conversation.

> to try their best to guess

Or perhaps ask for clarification.

Re: The End of OS X

#55

I know this is about the coming MacOS 11.0, but Catalina was the end of OS X for me, and the end of Macs in general for me. I will not be migrating from Mojave to Catalina, and will use my current generation of Macs until they are obsolete, while finally making the transition to Linux. Mac and I had a good run, but I'm clearly no longer the target demographic and their corporate goals aren't aligned with mine as a cu…

> I will not be migrating from Mojave to Catalina, and will use my current generation of Macs until they are obsolete, while finally making the transition to Linux

I've recently been doing this, after some time (ahem, a decade) away from "serious" desktop Linux. Some observations/pointers:

- Running ultralight window managers with minimal services still feels the most "right" and predictable/stable way to have a Linux GUI. So that hasn't changed. Unfortunately this means manually screwing around with scaling and size problems on 4K displays now. Like, really weird stuff like various windows being drawn at different scales or part-scaled-part-not and drawing your cursor a different size when you mouseover. It's no fun sorting all that out. Boo.

- Wayland is less horrible than I expected but does give the impression of being a lovely new way for any ordinary application to crash your entire windowing system. As if Linux didn't have enough of those already.

- It's still easier and more predictable to just run the damn thing in a Virtualbox VM under Windows, if you've got the horsepower for it, letting Windows handle the drivers for the actual-real-hardware. This may not be true on stuff like System76 or oldish Lenovo machines or something, but probably is if you've got, say, a dedicated recent-vintage graphics card, unusual USB devices, anything like that. It's just way less crashy in a VM, on the same hardware, unless you're damn lucky and probably running fairly old kit. Was true in 2010. Still is, it seems.

- Gnome3 has gotten so incredibly resource-hungry and unstable that it's finally driven me to KDE (see above about handling scaling issues in the lighter WMs being a PITA for why I'm not in Awesome or XFCE or something), which is a DE I've never liked since I first tried it two decades ago. It is a little better now, admittedly, and doesn't make me feel ashamed or like I'm Doing It Wrong for not running exclusively K* applications like it used to. The settings panels are still some kind of bad joke, organizationally. It's not that there are too many options, it's that they're trollishly organized. Friggin' gnome-shell. What an absolute garbage fire. Which is a shame because aside from being even harder to customize meaningfully than macOS (!) Gnome3 looks fairly pretty. It's seriously, no joke, awful though. I wish I could use it but it's simply broken, at present.

- All the new packaging stuff I've tried sucks. AppImage is closest to being good but it's exactly as good as having a bare exe file on Windows or downloading a static binary on Linux, so... it's fine, but does nothing for you like adding console launch commands ("code ." for example) or configuring your environment, as far as I can tell. None are as good as Homebrew and Homebrew-Cask (yes it's on Linux but there are way fewer eyeballs on it so fewer packages and they're often broken) for managing user-, not system-level packages. Snap's complete crap, like, they should just give up, it seems irredeemable at this point. Flatpack is significantly better but still not great. Usable for some unimportant things if you don't mind them doing weird, broken stuff sometimes or just not working when you need them to (see again: unimportant things)

Incidentally, I've settled (back) on Debian after trying Fedora 32 due to tons of recommendations on Reddit (I... did not enjoy the experience) and already knowing I didn't care for Ubuntu (anymore—it was great in the late 00s) as I'd tried that in VMs a few times during my Linux hiatus and not liked it. I also toyed with Void Linux on a crappy little machine for a while and damn is it nice for low-end hardware. I tried Manjaro but I-forget-what serious thing was broken about it on my rig and was going to be a pain to fix. Great installer, though. 10/10 on that.

Re: The End of OS X

#56

OS X was a marketing name and was dead for a while now. Besides the CPU arch change and a slightly refined UI it is pretty much the same.

Homebrew packages work fine, you can still disable SIP, Gatekeeper hasn't gotten tighter. Everything is almost exactly the same as Catalina. There's really no substance whatsoever to the claims that MacOS 11 is locking the OS down. You could make this claim for Catalina, but Big Sur isn't changing anything of consequence at all.

I see some smart decisions here then. They don't have to admit to the issues with the changes they made last time. Apple would probably argue that developers haven't moved quickly enough in the right direction while those outside would argue differently.

It'll be interesting to see what things look like later since this architecture transition will weed out some problems Apple see, but in a different way. On the other side of the transition they can make further changes to lock down the OS with fewer issues. This has to be Apple's ultimate goal because it's what serves their needs best from many perspectives. I wouldn't suggest it serves the needs of many devs though.

Re: The End of OS X

#57

Earlier quoted context omitted.

The comment states the why clearly when read within the context of the article: "I'm clearly no longer the target demographic and their corporate goals aren't aligned with mine as a customer any more"

Did we miss a memo where Apple suddenly changed their target demographic and corporate goals? I don't think they did. The comment seems like an emotional nostalgic conclusion they have after watching the WWDC, without any mentions why they think future will be worse than what we have now or what we had in the past.

Those things are rarely (publicly) unambiguously stated, so no, we didn't miss the memo because there wasn't one. Does not mean that over the course of time a target demo can change for a company though. Apple would hardly be the only company who did.

Pity though, Snow Leopard was something to be envious of.

Re: The End of OS X

#58

What I think a lot of us programmers miss is the experience of the end-users who are likely to buy a Mac. The fact of the matter is, 99.9% of people who buy a Mac are not going to know what subpixel hinting is. Or care too much that they can't delete Chess.app. To them, the Mac is nicer looking and easier to use than ever. Will iOS devs care? Not at all - developing for the Mac has become easier with Catalyst. It's e…

Overheating is still a concern for me.

Re: The End of OS X

#59
> an open platform on top of the tremendous hardware innovation being driven by the iPhone sounds amazing.

macOS is anything but open though. If anything, it's completely the opposite, in many senses of being not open. If you want an open platform, use Linux.

Re: The End of OS X

#60

> what the company perceives as a weakness: the paucity of apps in the Mac App Store. I think that the issue with the Mac App store, is that it imposes restrictions on a once-free-for-all development environment. I know many app developers that have abandoned it, after giving it a damn fair shake. Mac developers just got used to being able to wander all over the house, poking their noses into whatever they liked. iOS…

>the iOS model is the one that needs to be followed, if security and privacy are the goals

How so? Why should a centralized store be the zenith of these goals? Even if you assume the only way to be protected is for Apple to anoint binaries as good and pure, you would still gain privacy through decentralized distribution and secure signatures even if Apple controls the whitelist.

Post reply on HN