Darling – macOS Translation Layer for Linux • Hacker News Board

Re: Darling – macOS Translation Layer for Linux

#71

Post by michaelmrose » Mon, Apr 29, 2019, 5:40 AM UTC

msbarnett wrote: post #58

techntoke wrote: post #56

Earlier quoted context omitted.

> System-wide UI consistency, especially in the fine details > Singular UI/UX vision eliminates points of confusion and everything-is-a-compromise choices for third party developers There are consistent themes across both GTK+ and Qt. Pretty much all my apps share a similar theme and UI. There are Mac Apps as well that deviate from whatever "standardization" you're referring to. > Navigable by non-experts, even when…

> There are consistent themes across both GTK+ and Qt. And then you open an app that’s neither GTK+ nor Qt, and you’re back to square one. Besides, there’s more to UI than “theme”. I have ^w mapped to delete word on OS X. One line in one config file. It works universally in every text box on the system. I tried setting that up in Linux. Eventually got it working through some gnome setting or some such. In some apps.…

The majority of users use virtually no hotkeys. A truly shocking portion don't even know how to copy and paste with their keyboard. Weep when you watch people who don't even know how shift keys work and hit caps lock letter caps lock.

I don't honestly believe that the inconsistency in colors or gui elements that exist on a typical linux system are a meaningful barrier to anyone.

A typical desktop ships with mostly qt or mostly gtk apps that share the same look and feel + a browser that looks and works like the user is used to working across platforms.

In a sense the browser is actually the single most important app for the majority of users and its more useful for it to be consistent with expectations than consistent with the desktop.

I think power users are more apt to be put off by small differences in much used keybindings but they are the minority.

While consistency may seem like an impossible task there are really 4 really big camps. Chrome Firefox gnome kde.

This seems like a small enough group that getting everyone to agree on a common way to communicate to all the desired way to define keybindings ought to be tractable.

I think this is a worthwhile endeavor we ought to persue.

Re: Darling – macOS Translation Layer for Linux

#72

Post by michaelmrose » Mon, Apr 29, 2019, 5:43 AM UTC

msbarnett wrote: post #70

flukus wrote: post #62

Earlier quoted context omitted.

> And then you open an app that’s neither GTK+ nor Qt, and you’re back to square one. The same is true on every OS, if you open a java app on macos it's not going to look right, do you blame apple for that or the developers of the app? > In some apps. Firefox didn’t respect it and wanted its own setting iirc. Will firefox respect it on macos? It's not a good example anyway because firefox is not native, firefox is th…

> The same is true on every OS, if you open a java app on macos it's not going to look right, do you blame apple for that or the developers of the app? I’d blame the app, because in the case of OS X there is a single consistent set of UI components that effectively everything uses, and being gratuitously incompatible is on the app. In the case of Linux there is only a hodge-podge bazaar of gratuitously incompatible U…

There are really mostly 2. Gtk and QT and themed with similar themes they aren't different enough to be too meaningful.

Re: Darling – macOS Translation Layer for Linux

#73

Post by ncmncm » Mon, Apr 29, 2019, 5:49 AM UTC

The main attraction of Apple products is the certainty that, no matter what you do, they cannot be made to do what you would like. The consequence is that you feel no urge to spend time tinkering to get there, and instead adapt yourself to what it actually does. This reclaims all the time and attention that you would have spent on tinkering.

Since what Apple products do has proven adequate for a large number of people, and you are no less adaptable than they are, you know you can adapt, too.

Having once chosen to adapt to what Apple has chosen to offer, you find it easier each time, until it becomes wholly unconscious. Each time Apple takes away something you had used, you might momentarily balk at the "upgrade", but always acquiesce in the end.

[Edit] So, the appeal of Mac emulation is very limited, because it starts out with tinkering.

Re: Darling – macOS Translation Layer for Linux

#74

Post by michaelmrose » Mon, Apr 29, 2019, 5:49 AM UTC

msbarnett wrote: post #70

flukus wrote: post #62

Earlier quoted context omitted.

> And then you open an app that’s neither GTK+ nor Qt, and you’re back to square one. The same is true on every OS, if you open a java app on macos it's not going to look right, do you blame apple for that or the developers of the app? > In some apps. Firefox didn’t respect it and wanted its own setting iirc. Will firefox respect it on macos? It's not a good example anyway because firefox is not native, firefox is th…

> The same is true on every OS, if you open a java app on macos it's not going to look right, do you blame apple for that or the developers of the app? I’d blame the app, because in the case of OS X there is a single consistent set of UI components that effectively everything uses, and being gratuitously incompatible is on the app. In the case of Linux there is only a hodge-podge bazaar of gratuitously incompatible U…

Jetbrains had over a quarter of a million paying users in 2017. I'm not sure how much the average paying user spends but presumably they make somewhere between 37M and 125M gross. More money means more polish. Logically we need to invest in what is important to us.

Re: Darling – macOS Translation Layer for Linux

#75

Post by michaelmrose » Mon, Apr 29, 2019, 5:56 AM UTC

MuffinFlavored wrote: post #3

peter_d_sherman wrote: post #2

Something like this is sorely needed by the open-source, not-for-profit, shoestring-resources, cross-platform development community... For example, let's say that I'm writing some cross-platform open source software. And let's say that I am developing this software on a Windows or Linux Box. Now, let's say that for whatever reason, I don't have and can't afford a Mac (a scenario like this is more common than you woul…

> Now, let's say that for whatever reason, I don't have and can't afford a Mac. Rely on some developer(s) spend 10,000 hours of their time for free developing a Mac emulation layer, or go on eBay and buy a $600 used Mac to test on...

The logical option c is just to not support mac unless your target market is one disproportionately likely to use macs like developers or artists in the US.

Re: Darling – macOS Translation Layer for Linux

#76

Post by zapzupnz » Mon, Apr 29, 2019, 5:56 AM UTC

mycall wrote: post #68

zapzupnz wrote: post #38

Earlier quoted context omitted.

I'm with you on that, and it seems to be symptomatic of the project's refusal to step away from the legacy of mimicking OPENSTEP. They've made 1995 look like a goal towards which to aspire. They've added theming support to GNUSTEP, but that's really lipstick on a pig. Then again, I can't blame them too much, the project is mostly quiet and doesn't attract as much attention as the other popular Linux toolkits.

With material design and fluent design, OPENSTEP is fossilizing.

GNUSTEP, you mean. OPENSTEP is the fossil GNUSTEP wants to be.

In fairness, they've made strides since then. The Foundation library seems to be aiming for compatibility with a macOS release from several years ago, at least still more recent than the early-to-mid 90s.

Re: Darling – macOS Translation Layer for Linux

#77

Post by michaelmrose » Mon, Apr 29, 2019, 5:59 AM UTC

earenndil wrote: post #45

zapzupnz wrote: post #5

Earlier quoted context omitted.

> For example, let's say that I'm writing some cross-platform open source software > OK, so now how do I compile/run/test the Mac version of my software -- without having that Mac? You don't. Darling is not a suitable replacement for testing on a real Mac in exactly the same way WINE is not a suitable replacement for testing Windows software. A developer who thinks so is misguided. Virtualising macOS would be more su…

> You don't. Darling is not a suitable replacement for testing on a real Mac in exactly the same way WINE is not a suitable replacement for testing Windows software. Disagree. Darling is not quite there (yet), but wine could reasonably be used to test dev builds, so long as a native windows version was trialed before release.

I tried this with a c++ class surely I can just recompile the final project somewhere towards the end of the class and everything will totally work!

It didn't go well.

Re: Darling – macOS Translation Layer for Linux

#78

Post by michaelmrose » Mon, Apr 29, 2019, 6:01 AM UTC

swiley wrote: post #27

lostgame wrote: post #15

I wish I had the time to contribute to this. Getting Logic running on linux would let me ditch the Apple hardware. It's usefulness is entirely limited to the implementation of the core library frameworks such as Core Image, Audio, MIDI, Animation, Data, etc...which will be very, very difficult, I think, while maintaining FOSS status.

Linux audio latency really isn't great, having another layer in between your app and the kernel probably won't help.

Its actually fine and has been for a long long while. Pulseaudio kind of sucked when it first came out.

Re: Darling – macOS Translation Layer for Linux

#79

Post by themodelplumber » Mon, Apr 29, 2019, 6:14 AM UTC

ncmncm wrote: post #73

The main attraction of Apple products is the certainty that, no matter what you do, they cannot be made to do what you would like. The consequence is that you feel no urge to spend time tinkering to get there, and instead adapt yourself to what it actually does. This reclaims all the time and attention that you would have spent on tinkering. Since what Apple products do has proven adequate for a large number of peopl…

I tried the adjustment to the Apple way from 2010 to 2018. By the end of it I realized that "tinkering" has an equivalent in the Apple world and it is called "upgrading." These two terms are almost synonymous in the pain which they will cause. By the end of my journey I had a firm "one major version upgrade per iOS device" rule and a desktop OS upgrade experience that was driving me crazy, because random CAD app wouldn't support the new OS, but another app wouldn't support the mainstream-old OS anymore.

I won't say that tinkering is totally productive, but 1) personally my tinkering on Linux has led to technology development and I'm the type of guy who benefits from that anyway, and 2) I believe the pain you leave behind in lost productivity is at least offset by 25% instant upgrade-related stress for each Apple device in your household or sphere of personal work activity.

Even just constrained to hardware, the horror stories about new Apple devices alone made me doubt my HW upgrade plans. By itself that was enough to make me wonder if I was about to throw away thousands.

Unsurprisingly I felt like there were things Apple could do to make this all better, but like you said, they cannot be made to do what you like :-)

Re: Darling – macOS Translation Layer for Linux

#80

Post by vaylian » Mon, Apr 29, 2019, 6:21 AM UTC

ncmncm wrote: post #73

The main attraction of Apple products is the certainty that, no matter what you do, they cannot be made to do what you would like. The consequence is that you feel no urge to spend time tinkering to get there, and instead adapt yourself to what it actually does. This reclaims all the time and attention that you would have spent on tinkering. Since what Apple products do has proven adequate for a large number of peopl…

Thanks for putting into words what I had suspected for so long. Apple is really good at identifying what the vast majority of people (aka normal users) want to do with their computer and then they optimize these workflows. But if you want to do something more exotic you are usually out of luck.

Talking to Apple enthusiasts is really tough. It's almost as you are speaking a different language. Of course, in the end of the day a computer is just a tool. They are happy with what they can do with their machine and I am happy with my machine. But they have a hard time understanding why I see their system as limiting.