For the majority of users, this is irrelevant because linux is unusable for them. And they are the customers. To have a sustainable business, we have to target the platforms that paying customers use. That platform is not gnu/linux.
I got Ubuntu recommended by an electrician back in 2005 or 2006. (I used to be using Red Hat/Mandrake/Mandriva back then.)
Soon after I realized it had come to the point where Ubuntu was not only possible to install and use but simpler to install and use than at least OEM-loaded Windows machines.
It seemed to work well for grandmothers, kids and sysadmins.
Notable exceptions:
- certain people refuse to learn (certain Windows sysadmins)
- people who depend on powerhungry software that doesn't exist on Linux (for less power hungry software there is RDP or Citrix and Windows Terminal servers.)
The real problem for me is that support is really hard to come by if something goes wrong. Grandma can't access the internet on her Ubuntu system — who can help her? I'm 500 miles away, my parents are equally as far in the other direction. Neither myself or my parents use Linux as a daily driver either. Her neighbors have Windows or Mac machines. Does she drive to Best Buy with her desktop? Call in external help?
If she uses Mac or Windows machines, help is just a lot closer. My dad can replicate her issues on his machine and walk her through it on the phone, or I can. Her neighbor will at least potentially be familiar with the way the OS looks, so can offer greater help.
I'm not sure how you can close that gap if you're not next door to the individuals who would benefit most from it. While it may be a great individual solution, the support burden just isn't a solved problem in my opinion.
ubuntu (and i'm sure others) are in a great state for casual users already -- i've handed multiple family members an ubuntu machine with no more instruction than "use it like any other computer, don't be scared by the icons looking a little different" with great success. for the increasing number of people who do almost all of their work in a word processor and a browser exclusively, it's a complete solution with almost zero hassle.
There is no such thing as to much RAM
There is no such thing as to much DASD (Disk)
There is no such thing as to much Screen Size
There is unfortunately never enough Budget :-)
Try to tell Red Hat/Canonical/SUSE about it.
The biggest OP mistake is relying on iOS only. Spending 9 years as iOS developer - great, but if you're mobile dev working with two platforms will be a life savior for cases like this one.
> For the majority of users, this is irrelevant because linux is unusable for them.
While I think I get what you're trying to say, I don't think it's necessarily unusable for regular users, more so unreachable in a nice, preinstalled form from a well-known brand & retailer.
Also, there's no ads for GNU/Linux machines on TV or on the Metro, which I think is often underappreciated.
I know because I've personally witnessed friends and family members who swore Linux is unusable give it a serious try only after Windows or macOS crossed them for the nth time and haven't had a single one switch back.
The key is to get hardware that is well-supported and not 2nd tier i.e. not a netbook etc. that I'm just going to throw Linux on and compare the experience to my 10x more expensive MBP.
The experience's not perfect, but that's not the case for macOS either. Is it "unusable"? Far from it.
The key is to get top-class hardware with the express intention of it running Linux, rather than installing it on the laptop Windows will no longer run reasonably on.
It is true however that such as switch is easier for developers like myself who are not dependent on Cocoa/SwiftUI etc. but that's a choice you have to make.
Linux does run better than Windows tends to on old HW, sure, but comparing that experience to macOS on new hardware is less than fair.
There's plenty of new hardware Linux runs great on, I am not sure what you call "experimental" hardware, but I am using a Surface Pro 7 as my main machine with exclusively Linux on it, no problems.
Any recent standard ultrabook should also work fine, heck I even have a random, no-name Chinese UMPC that works well.
Much as I loathe Windows and increasingly dislike MacOS, there is nothing like Adobe CC or any of the major DAWs, video editing tools, or VJ performance suites on Linux. Not even close.
There are toy copy apps run as hobby coding projects. But professionals need the real deal - for file exchange and for other workflow reasons - and Linux simply does not offer that.
If your desktop product is good enough, your users will install Linux. Linux runs on Windows and Mac.
Your mobile app can be a web app.
Not a GUI app per se, but an app used on a desktop.
2) Taking his argument in good faith, he also means "free" apps with microtransactions, because in both cases the same situation exists: The iOS App Store is too big a market to ignore. Which leads to the same problem, regardless of whether your app is paid or "free", that if Apple suddenly decides it doesn't like you, you're screwed.
What I've noticed on macOS however, is that is a lot easier to sell things that should be bundled or FLOSS for relatively serious money. For example there are dozens of rather expensive "Finder replacements" on macOS and you'll have a hard time selling something proprietary like that to Linux users because we do have good file managers that are libre software already.
- sound editing/musical production
- video editing/special effects
- CAD/CAM/EDA
- Mechanical/physics/electrical simulations
- interfacing with custom hardware
Hollywood studios run on Linux (except Pixar, and granted, not with webapps), so it is possible. What's missing is the middle-ground, mass-market solution like Premiere/After Effects or FCP, though DaVinci Resolve is quite nice entry there.
- interfacing with custom hardware
From experience, that's a lot easier on Linux.
As for CAD, video editing and music production - sure you do have a point, except that in those areas the likes of Autodesk and Adobe dominate anyway and the "pro" segment of macOS/Windows users is a tiny fraction of the overall userbase, so I'd be rather surprised if that's the kind of software you can sell in decent numbers to macOS/Windows users as an indie dev.
I'm very lazy and I don't want to tinker with shit to get it running or maintain it. So, I stuck with Windows for the longest time. It was the easy path for a long time because a lot of my work was doing .NET. However, once I switched full-time to doing Node.js Windows got in the way more and more. I spent a lot of time investigating issues. I have a couple of Macs and I tried using macOS as my main, but honestly I can't stand anything about the way Apple does things. I just hate the bad window management, the shitty finder, the stupid global menu bar and the lack of a coherent hotkey system.
I had been trying out Linux desktop distros for honestly decades, out of pure interest. Every single one of my experiments ended in a failure to boot one day after an update. Then one day I read about Manjaro and I decided to give it a shot. 2 years later, I'm still running it as my main work OS on 3 different workstations (home desktop, work desktop and a laptop). Not one of them has ever eaten itself due to a bad update. I've had to fix minor issues after a couple of updates, but that's nothing compared to the shit I've gone through to make Windows or macOS usable.
If you're new to Linux, skip Debian, Ubuntu and all that crap and go straight to Manjaro. Installing software in Manjaro is way, way, way easier than any other distro and the stability for me on 2 desktops and a laptop has been nothing less than stellar.
