'Next/Apple' isn't a quick playbook - it's an over 30 year R&D effort to create a hugely complex software and hardware business, and it spent about $100 billion in R&D to get its products where they are today (at the absolute cutting edge of technology). Writing your own modern OS and building/manufacturing good hardware to compete with this is difficult enough, and then you have the even bigger challenge of getting all the major software vendors to support your new platform.
But isn't this much, much easier if you just piggyback on the Apple hardware ?
I always expected this to happen.
Circa 2008 or 2009 I thought that any day now there would be a linux distribution built specifically foe one single Apple laptop. No hardware issues, no gremlins, no moving targets - you would have a (very) fixed hardware target and optimize just for that. Then I, as a user, could just go to the Apple store and buy a nice shiny device and install MBAlinux on it and call it a day.
I really don't understand why this never happened. Further, in many ways it seems that the opposite of this happened - installing linux/FreeBSD is weirdly painful on Apple laptops which is unexpected since we all know what is inside of them and the installed base is huge.
So I would suggest that you could, indeed, build a hardware/software ecosystem - just let Apple build the hardware part ...
Sure, it's easier, but then I'm not sure what the point is or what makes it one of the biggest opportunities of our time.
I also understand why it never happened - there is already a unix-based OS which is designed with perfect compatibility with the Apple Hardware called OSX! I'm not sure what the advantage to a consumer would be for replacing OSX with linux - other than the fact that it gives consumers choice - but of course providing a distro that only operates on a specific Mac is then limiting hardware choice so it doesn't really solve that in some respects.
And if it's just for developers, then wouldn't developers want some choice of hardware, good support for tooling, the ability to test native apps without virtualisation e.t.c.
IMO I suspect the Venn-diagram of developers who:
* want a Mac but don't want OSX
* don't mind that they can't upgrade their hardware
* are willing to run some totally-new operating system
* Accept that it will initially lack the support of the runtimes they use, and some software, and won't be able to develop certain types of software because of this.
* Accept that if they wish to continue using the OS for their next laptop they will be fully locked-in to a single hardware model.
is pretty vanishingly small.
Lets take Ubuntu as an example. Today you can get Ubuntu laptops that will work out of the box. Is that true tomorrow? Absolutely not. The next distro version will break something in the hardware. I have been burned by this twice now. At then end of the day the Apple premium is not really a premium. It ensures that they continue to support their legacy hardware for years. The people who bash the premium as some sort of "idiot tax" are actually valuing the software that runs on the machine at 0$. There are too many people in this world that don't understand how much effort it takes to create and maintain good reliable software. You see it on the app store where people can't fathom spending 99 cents and you see it in the bashing of Apple devices.
Lets assume that your hardware works beautifully with the current version. Then you actually look at the apps shipped with the distro. They are poorly made and do not form a cohesive OS. You are forced to hunt for other open source equivalents to basic stuff like "paint". Have you tried using the calculator or notepad equivalents? They suck compared the simple and easy to use Windows and Mac equivalents. This is something even Windows gets right. It comes from the fact that Canonical does not have the resources to build each app around a unified design and UX principle so they farm it out to the "open source community".
Finally, why do each distro version seem to break something on the same hardware year after year? There seems to be a serious lack of regression testing on these distros. For 10+ years I have witnessed how one version of Ubuntu breaks some stuff, fixes others and then the next version fixes some stuff but breaks previous working items. Then it gets worse, the subsequent version breaks previously fixed stuff again! I am forced to QA the entire OS every time a new release comes out and hope I don't miss something(which I always do)!
I did try and use Linux full time but the UI drove me up the wall so I’m back comfortable in macOS but my dev work flies on that Linux box.
Isn't it a bit on the nose that you accuse Linux of failing at the one thing that Windows is notoriously bad at, UX cohesion?
More seriously -- Linux reflects a different mentality and way of doing things. It is not for everyone. Downloading the software you want is the expected way to do things. I have no idea whether Ubuntu ships a "paint" replacement, but regardless, Pinta or Paint.NET are like three clicks away, thanks to the software repository approach.
Linux is more or less for people who want to experiment and configure things their own way, and make software that solves their own problems in the way that they want those problems solved. Creating a single, opinionated, out-of-the-box working desktop experience with perfect hardware compatibility with whatever bullshit proprietary-blob using silicon is out there is (a) hard, and (b) not what most Linux-using developers are interested in.
The people who use Linux largely recognize that yes, it is a compromise, but also that using Windows or macOS also represents a compromise. Having used Linux for nearly 15 years now myself, I can say confidently that the trade-offs for me weigh heavily in favor of Linux.
Yeah you can criticize Windows for trying to update their designs with Metro and the like but in reality, all the old apps that worked cohesively are still there even today. Ubuntu and the Gnome or KDE based distros never had this to begin with. Just multiple flavors of the same cruddy base applications since all the distro are using the same apps anyway.
>More seriously -- Linux reflects a different mentality and way of doing things. It is not for everyone. Downloading the software you want is the expected way to do things. I have no idea whether Ubuntu ships a "paint" replacement, but regardless, Pinta or Paint.NET are like three clicks away, thanks to the software repository approach.
Yeah thats fine but that unfortunately makes it a non-starter if you are looking for a direct replacement for macOS or Windows.
Yes Paint.NET/Pinta/GIMP are always trotted out when I post this example. Pinta has been an unstable mess every time I have installed it. Plus "paint" is a near-instant loading app that is several MB in size whereas Pinta is installing loads of supporting libraries because it is a more complex application. Your telling me that in 2021 they can't just ship a simple app to allow a user to just jump in and use to resize images or add some text to basic images? This hinders the usability of the system when I can't just quickly do a simple task and move on! It is as if the developers of this distros have never understood how a regular user uses a PC.
>The people who use Linux largely recognize that yes, it is a compromise, but also that using Windows or macOS also represents a compromise. Having used Linux for nearly 15 years now myself, I can say confidently that the trade-offs for me weigh heavily in favor of Linux.
The only thing that is a given is that any comment bashing Linux will ultimately attract someone like you that tries to twist and turn my words to justify it. I've seen it for 10+ years now without fail so i'll leave it at that.
Its a shame because I have looked at the messy bug tracker for Ubuntu and have tried to fix issues but then I stop and realize what is the point when it breaks again in some subsequent version of the distro. I wish someone would just dump a bunch of money, hire former Windows/Mac devs and properly build a lot of the supporting components of some distro, then all the other distros can roll up those better apps and then we have at least something that can be called adequate in 2021.
The only thing that's a given is that any post that proclaims the relative merits of Linux versus alternative operating systems will immediately attract posts like yours bashing it, so shrug.
> I wish someone would just dump a bunch of money, hire former Windows/Mac devs and properly build a lot of the supporting components of some distro, then all the other distros can roll up those better apps and then we have at least something that can be called adequate in 2021
My point is not that you're wrong to feel this way, but rather that you should recognize that "adequate" is ultimately subjective. Adequate for whom? Adequate how?
The Linux ecosystem is largely designed by and for people who are willing to tinker, willing to customize, who want to design software that scratches their own itches, and who aren't looking for a perfect out-of-the-box experience from a distro. A handful of people want to bring about "the year of Linux on the desktop", but they're a minority and even for them the interest is usually secondary to their own use of Linux.
There's no "twisting your words" required here. What you want is a near-perfect out of the box Linux experience. What most Linux users want is ... something else. My point is simply that that's okay. Linux doesn't have to be for everyone. Your problems with it are not everyone's problems with it. In particular,
> that unfortunately makes it a non-starter if you are looking for a direct replacement for macOS or Windows.
Most Linux users don't want a direct replacement for macOS or Windows. Maybe there's a class of "theoretical Linux switchers" out there who would switch and would be the majority of Linux users if they did, but they are not, at present, the majority of the people using and working on Linux.
What Linux provides me with is (a) a well-integrated package manager containing fully free/libre software, (b) a comprehensible system (where I can understand fully how each part works), and (c) a modifiable system (where I can change how the system operates to the extent I want). Having a perfect replacement of the MS Paint application is not even on my radar. But that said:
> Plus "paint" is a near-instant loading app that is several MB in size whereas Pinta is installing loads of supporting libraries because it is a more complex application.
Maybe you're exaggerating, but on my system Pinta has an installed size of only 2.88 MiB and has only two direct dependencies. Maybe you're thinking of the fact that it's written in Mono (the C# runtime), but that's a shared installation with all other Mono applications. It's equivalent to Windows shipping with the .NET runtime or UWP.
I only ever need a laptop when traveling, I have a big desktop setup at home. I plan to take my Steam Deck traveling with a portable monitor and keyboard.
Apple doesn’t prioritize lack of binary blobs. The EFI firmware is all proprietary. All their Wi-Fi have been switched to Broadcom.
They do weird non-standard things to the Thunderbolt controller, e.g., you have to lie to the firmware and claim to be macOS in order for it not to disable the Thunderbolt controller.
Newer MacBooks hide a bunch of hardware behind the proprietary T2, and whatever embedded OS runs the Touch Bar.
MacBooks are not ideologically pure, and sunk efforts to get an OS working on other machines are often wasted on MacBooks because Apple does things in bizarrely different ways.
Would that be even legal? I mean selling a commercial OS that would be marketed to install as a replacement OS on the most locked, most proprietary hardware on the market?
The problem with this is which developers? People who write embedded systems? Web developers? People who write custom Windows applications?
Any given developer subset is likely to find this hypothetical new developer computer to be either too complex to use or not differentiated enough from Windows or MacOs (or ChromeOS).
> Most of the software we use to create software are electron apps
This is not true for most people whose primary employment is writing software, or working on software teams. Most people who get paid to write code work primarily in either the Java or .NET ecosystems and use something like Eclipse, IntelliJ, or Visual Studio. (Many more are using niche-specific tools in a captive platform like Oracle, SalesForce, SAP, etc.) If the new platform doesn't have 100.0% binary compatibility with legacy tools written for Windows and/or MacOS, its addressable market shrinks substantially.
VSCode is not the entire stack needed to build Windows desktop applications. There are tens or hundreds of thousands of developers who build applications for the Windows desktop. I'm not (for the most part) a .NET dev, but my current understanding is that only Visual Studio running on Windows is a first-class citizen with the ability to access all parts of the dev stack. The Windows dev stack doesn't need to move away from native Windows applications any more than does Xcode need to move away from MacOS.
It's critical to have a great developer story yes, but to make a stellar platform that needs to be balanced with a great user story, and that means developers might not always get everything they want down to the letter.
Why would that be an opportunity though?
It would be a low margin niche, with a small market segment, of which most would stick with Apple/Lenovo/Dell.
If you wanted to create a new platform, your best option would be to go the other way and make an OS that was “just electron” and ran all the electron apps in the world faster and better than anything else. Unsurprisingly Google has tried this with Chromebooks, but their track record on consumer product development is so poor that perhaps they just didn’t execute well and someone else could pull it off.
Another challenge is that if you did that you gain wide software compatibility but you lose any obvious differentiator. The likely way to win would be if you could make a laptop that was “just as good at running web apps as your Mac, with just as much battery life and just as nice hardware” but somehow cost under $400 or so.
I actually wouldn’t be surprised if we see that coming out of Chinese OEMs in the next decade.
Screens are the most expensive things on a laptop, mobile and tablet.
So you are not going to get a good screen
If they need a photo manager, in my experience the most common application need after a web browser, then Digikam really cannot be beat.
I am a full time Linux user. And I'll probably support anyone who wants to try it until the day I die. I absolutely love it. But we still can't enjoy some of the simplest use cases without screwing around with configs and in some cases, writing scripts that listen to DBus or udev.... So every time I hear someone say, "just use Linux" I think... nah, just buy a Chromebook (and - yeah, use Linux). If your needs are any more than that, Linux might not be for you.
Sadly, this is now true for Macs as well. Were by "not work" I mean: not finding/not supporting the proper resolution and/or refresh rate for the display.
I never succeeded. I just use the Macbook with only the built-in display now.
> writing scripts that listen to DBus or udev
Exactly. Not to mention the preceding step of spending 30 minutes in forums to find someone else who has had this problem on a system with exactly the same motherboard so you don't try the things that didn't work for them.
I'm still 100% Linux on the server, headless Linux doesn't have nearly the warts as GUI Linux.
In the context of the original comment, though, this does narrow down the market that can be addressed by a company running the Next/Apple playbook.
It seems vastly cheaper to solve these issues than to start a new company to produce its own hardware and OS.
If you did start such a company and prove there was a market, then you're making a bet that Canonical (or KDE devs) won't put you out of business.
> Want to use bluetooth? It might work.
Sounds like the problem of installing Linux on a hardware designed for Windows. All those things work flawlessly on my Purism Librem 15, which came with preinstalled Linux. (Ok, I did not try Bluetooth, but saw reports that it works.)
Here is one. Randomly, based on no relevant input from me or changes in the laptop's state, my network connection dropped and the Network Manager UI was telling me no network adaptor could be detected.
Some fumbling around in the Terminal (including various reboots not solving the issue), and managed to enable the wireless adaptor which apparently could be detected and connect to my network, though at the same time the UI was telling me in no uncertain terms that no wireless adaptor was connected to my laptop.
Then later, again randomly, based on no relevant input from me or changes in the laptop's state, the UI agrees there is a wireless adaptor connected after all. This is on a machine currently in near-factory state with certified compatible Ubuntu preinstalled.
I share this example because one can at least comprehend why random monitors or graphics cards or what not do not cooperate without fiddling, can comprehend certain apps failing and crashing, can comprehend other unusual bugs. The UI thinking and acting like there is not a network card for no reason whatsoever, on the other hand, is completely illegible to even competent users.
Someone needs to just commercialise a proprietary and at least initially closed version of Linux (so as to to turn a profit) with good design principles in mind and deal with lawsuits and license issues later. There is plenty of money in it.
For how much more would you siphon money into basically empty stores, to see it as having gone "100%" of the way?
Which is even more damning of Microsoft / Google management.
The sunk cost fallacy is not really about what the venture would actually do. One could be said to have fallen prey to it even if they double down and the venture eventually succeeds.
What's important is that at the time of the decision (a) the path doesn't seem to be working, and (b) they think "but I've spend too much to quit now".
This is more likely when one assumes it can still has a chance to succeed, than if they assume it will inevitably result in failure. Nobody that assumes inevitable failure would decide to continue.
All I know is at this point Microsoft had two options: continue investing into creating an alternative to Apple, or cancel their plans and sit back and let the Office/Azure revenue flow in.
Maybe it was a long shot, maybe they decided the size of Apple's customer base divided by two was not enough to satiate them, but whatever the case, they signaled that they do not have the talent/gumption/appetite for risk to pull it off. But if any company did have the opportunity to go for it, I would think Microsoft (and Google) with their income stream would have been in position to do it.
Both companies seem to dip their toes, but never follow through.
I would have liked to see it succeed if only for there to be more competition.
It also had a great 'copy and paste' feature
This hits hard
The prerequisite of being a "large tech company" to pull it off is already a huge barrier to entry.
As for the barrier, it is in the software and app ecosystem.
My point with the software is that most software people use to create software/apps is increasingly created with web technologies / electron apps, so the native app ecosystem on a desktop machine is increasingly weakening as a moat.
- Unix-like, developing on Windows is pure torture
- Don't waste my time with configurations, drivers and other crap, I want a machine that I can be productive with out of the box. Take my money if you have to, but I don't want to edit Xorg.conf ever again.
You can talk about polish all day, but no machine that doesn't satisfy both is even close to appealing for a majority of developers in my experience.
You are right about the native app ecosystem being less of a blocker, but that's in line with my point.
And physical hardware kill switches? This almost seems too good to be true.
It will not (yet) replace your iPhone, although you can get pretty far on your own if you don't mind SSHing into such a device and messing with stuff on your own.
If it was focused on developers, it would by definition not be re-running the Apple playbook. Developers are much too small a market to justify that level of investment.
I'd be interested to know what "we" means in this sentence.
Do you mean your team?
Or developers in general?
If the latter, I am not sure what industry you work in, but I very much doubt the majority of developers operate in the environment you describe.
maybe.
it would be amazing to build systems that are are so beautiful that they inspire people like berners-lee and carmack to do their things.
Most developers I know are happy with their mac or windows spyware, or their rough edges linux rig.