Comment by faeriechangling
2 years ago
What's impressive to me is how long Microsoft has been playing defence against the Linux Desktop successfully - they were doing this back in the 90s.
2 years ago
What's impressive to me is how long Microsoft has been playing defence against the Linux Desktop successfully - they were doing this back in the 90s.
Not really. People don't care about libre software, they care about software that works, is supported, and they don't have to think about it. The Linux desktop appeals to developers and hackers, so 4% is probably the best they're ever going to do.
On the flip side, look how much Microsoft is losing to Linux in the server space. Consumers don't care what servers their favorite sites and services are running on: they don't have to manage it.
> The Linux desktop appeals to developers and hackers, so 4% is probably the best they're ever going to do.
Probably not. The body of the article reports 6.24% desktop Linux penetration already.
The 4% reported in the headline refers to browsers which report Linux in the user-agent string. That is more down to browser choice than kernel choice. Specifically, a popular Linux browser has chosen to omit "Linux" from its default user-agent claim, hence the discrepancy.
It may be that developers and hackers are drawn to certain browsers, but that is beyond the subject of Linux.
I would add on to the "don't care" to say that (in my experience) your typical computer user doesn't even understand what an OS is. With many people that I've interacted with they just equate the computer and OS as one thing and have no idea what OS they're using, much less that it's actually replaceable.
> Not really. People don't care about libre software, they care about software that works, is supported, and they don't have to think about it. The Linux desktop appeals to developers and hackers, so 4% is probably the best they're ever going to do.
Also, it shows the power of defaults and just showing up. Windows is as big as it is because Microsoft made the deals with OEMs 30+ years ago to ship their software as the default option. Macs would likely be an also-ran if Apple wasn't standing behind their hardware, both in terms of the hardware itself and the sales/support channel, as much as they are (just look at the relative success of their retail stores vs. the relative failure of Amazon, Google, Microsoft, etc., in the same space).
> care about software that works
I'm tech support for my wife's Win11 laptop, and this is rapidly becoming not the case. Microsoft is building bridges across its most.
Microsoft still wins in the end, they don't care if UNIX has won the server room, provided they still get the money for Azure OS, or Azure Sphere OS.
Looking at their earnings, it is going pretty well.
it's going pretty well right now, but if they keep massively increasing prices, it won't keep up forever.
the company i work at pays almost double now compared to a year ago, for arguably worse service. some of us in IT are looking at options, because being locked into a service that keeps shooting up in price is a huge risk.
yes it'll be hard to get away from the ecosystem. but with the increasing price, demand for alternatives surges too, and i can't see it taking too long until we get some options.
1 reply →
There are also people that prefer to use free / open-source software because of it giving power & control to the users & community, opposed to proprietary software where a single (typically commercial) company controls the software.
The disadvantage of propietary software is that you are at the mercy of how the company decides to maintain and develop their software. And they might do undesirable things towards their users. Like displaying adds, tracking / profiling their user's personal data. Changing the pricing. This typically happens when they gain enough market share and have enough users vedor-locked into their software.
And the vendor lock-in also makes it harder for users to make the software work with other 3rd party software. So generally propietary software isn't easy to combine with third party software, unless the company has a commercial incentive to support it. But often one of the reasons for the software to be propietary is because of the vendor lock-in. So that you will use all software from 1 company that all works well together, but doesn't work (as well) when trying to combine it with third party software.
> so 4% is probably the best they're ever going to do
Why would you say that? Take a look at the trend plot shown at the link. The trend started increasing markedly about two years ago. I wouldn't think it would suddenly flatline. At least I hope it doesn't.
https://en.wikipedia.org/wiki/Crossing_the_Chasm
> with a particular focus on the "chasm" or adoption gap that lies between early and mainstream markets.
> The author argues there is a chasm between the early adopters of the product (the technology enthusiasts and visionaries) and the early majority (the pragmatists)
Early adopters are a small minority, it's hard to give exact numbers but 5% sounds about right.
It's MUCH harder to get the pragmatists, late adopters, etc to use a new product. It generally requires a revolutionary product or brilliant PR and marketing (Linux doesn't really have PR and marketing and revolutionary would be a strong term for a product that's been in the market 30+ years).
2 replies →
> People don't care about libre software, they care about software that works, is supported, and they don't have to think about it.
Which long term libre software helps them do. Without it companies just keep stealing the open source features and moving the target of "good".
Sometimes you have to use very broad definition of "works". Often (certainly not always) UX and general aesthetic design for many open source apps is extremely poor compared to their proprietary alternatives.
They're playing defense vs ... is "linux desktop" even trying?
I don't expect someone out there to just make it work for everyone else, but outside specialized situations like gaming hardware, or phones, I don't see anyone really putting the work into appealing to the mass of windows users, at least not as far as covering all the bases that you would need to cover to really think about getting people to move.
MacOS is probably the most organized effort and there they are, and they don't cover all the windows features exactly.
That's not a knock on anyone here, just that they're maybe playing defense vs no no real organized effort.
> is "linux desktop" even trying
No. I've been using Linux since 2007 and spent most of my career so far in the Linux space. My goal has never been to convert Windows users to Linux. I don't care what others use so long as it doesn't impact me. I just want to make Linux better, because I am also a Linux user, and want my own experience to be better. If my work means someone has a pleasant experience and converts, then great, the more the merrier. But it's not my goal. So I think you're right, we're not trying to get people to move from Windows. We're just trying to make something we want to use.
> I don't care what others use so long as it doesn't impact me.
It does impact you regarding the depth of hardware support, for example.
6 replies →
As someone who has used windows for 20 years, and used linux for a combined total of about 3 years (including now since sept. 2022):
Linux desktop plays defense against itself. As long as linux is heavily CLI dependent to get stuff done, it's going to be stuck in sub 10% penetration. I don't see this changing either because the people who maintain linux love linux the way it is.
Don't get me wrong, linux is extremely powerful, but it's gate kept behind a fetishization of archaic and cryptic command line interfaces.
It’s not a fetish.
I think Lennox really opens up for windows users when they understand that everything is a Linux terminal command.
When you go into the GUI and change your display resolution, that is a terminal command. Etc.
This is the same for extremely popular things like VS Code, where all the extensions are super cool, most of them are also terminal commands or have some semblance.
However, I can agree with you that this distinction is never clearly explained. And the majority of windows users have no idea, so they come over to Lennox and don’t understand why things might look strange, or require a terminal.
It's not fetishization; it's an engineering OS.
Windows is a consumer OS, which is terrible for engineering, less centered around CLIs but nevertheless having a few even more cryptic ones tacked on like afterthoughts or a legacy hoarding exercise.
CLIs will always be more powerful than GUIs. They tap into what the OS actually is. The GUI is an illusion.
Windows actually has a very powerful CLI and really nice Terminal app. It's great for software engineering, I use it every day!
...and hence perpetual <5% adoption.
1 reply →
I'm not shocked. They've successfully "played defence" against a host of others, including OS/2, Be, and DR-DOS. And that's just the operating systems. Don't forget Stac, WordPerfect, Lotus, Novell, Sun, and a host of others.
The shocking bit is that Linux is even still alive, given the graveyard of Microsoft competitors.