Consumers may have left Windows for Linux somewhat but most are still stuck there as well.
I work in a STEM-heavy university. Nerds everywhere, and many are computer nerds. I only know one other person here that daily-drives linux out of the hundreds I work with. I am very sceptical of the consumer market share statistics.
I wonder if the statistics are skewed as “one machine, one vote” rather than “one person, one vote.” Because I know a few people (including myself, when I was younger) who would run like multiple Linux machines and start up random VMs to do a few things here and there.
Along side what you mention, I work for a company with a lot of engineers from multiple disciplines. Linux is everywhere in our environment and has been growing. Most of our engineers either have a Linux laptop, use Linux virtual machines or submit jobs through Linux backed software stacks. Our products have tools built on Linux to the point that we used to maintain our own Linux distribution to make having all those tools on a system easy, though we’ve just moved to having scripts to install and setup everything. Almost all of those engineers also have a Windows laptop as their daily driver. The few who don’t have Macs.
I don’t doubt that organizations can move to all non-Windows, but the Windows Ecosystem of Exchange, Office and Active Directory/Entra is very hard to replace. Yes, you can absolutely cobble together alternatives. But, you then have a special snowflake of an IT infrastructure. Maintenance may be eased by the use of tools like Ansible, but even that will likely result in organization specific playbooks and scripts. But centralized management of a Linux environment isn’t quite as standardized yet and so requires more planning and effort. And then there is the challenge of security.
Folks like to tout the inherent better security of Linux. As someone who works in Cybersecurity daily, I fully agree with that assessment. However, “better” isn’t “perfect”. You still need vulnerability management, monitoring and response tools. And this is an area where Linux lags behind Windows, mostly because of the effort put into it. Something as basic as Endpoint Detection and Response (EDR) on Linux is still a foreign concept to a lot of folks. And yes, it’s absolutely necessary. I’ve personally worked cases involving compromise of Linux systems and Linux specific malware. And more than just XMRig clone #5374. It may be harder to move from user to root on Linux than moving to local admin on Windows (unless some jack-off decided that Linux doesn’t need updates and DirtyCow still worked); but, there is still a lot of damage which can be done without root.
Overall, I’d say that the Linux ecosystem is evolving. It’s already embedded in a lot of environments of all sizes, but it’s usually a smaller part of a much larger Windows network. Hopefully, if enough large organizations start pushing away from Windows and towards Linux, there will be more investment and standardization. And, we’ve already seen some of that. Red Hat is a common distro of choice for large organizations because of that higher level of centralized control and standardization. Of course, a lot of the tools engineers want to use aren’t on Red Hat, specifically because of that centralization (read: everything AI). So, it will be interesting to see where we land.
There are a lot of other stats around, like statcounter, cloudflare, wikimedia, and even the data pornhub publishes from time to time. All of them show linux growing, and they all have different audiences and maybe methodologies
The article acknowledges the difficulty in gathering data, and moreover for their desktop numbers they reference StatCounter which itself only counts web-based usage on a subset of pages which are part of their network.
I work in a STEM-heavy university. Nerds everywhere, and many are computer nerds. I only know one other person here that daily-drives linux out of the hundreds I work with. I am very sceptical of the consumer market share statistics.
I wonder if the statistics are skewed as “one machine, one vote” rather than “one person, one vote.” Because I know a few people (including myself, when I was younger) who would run like multiple Linux machines and start up random VMs to do a few things here and there.
Along side what you mention, I work for a company with a lot of engineers from multiple disciplines. Linux is everywhere in our environment and has been growing. Most of our engineers either have a Linux laptop, use Linux virtual machines or submit jobs through Linux backed software stacks. Our products have tools built on Linux to the point that we used to maintain our own Linux distribution to make having all those tools on a system easy, though we’ve just moved to having scripts to install and setup everything. Almost all of those engineers also have a Windows laptop as their daily driver. The few who don’t have Macs.
I don’t doubt that organizations can move to all non-Windows, but the Windows Ecosystem of Exchange, Office and Active Directory/Entra is very hard to replace. Yes, you can absolutely cobble together alternatives. But, you then have a special snowflake of an IT infrastructure. Maintenance may be eased by the use of tools like Ansible, but even that will likely result in organization specific playbooks and scripts. But centralized management of a Linux environment isn’t quite as standardized yet and so requires more planning and effort. And then there is the challenge of security.
Folks like to tout the inherent better security of Linux. As someone who works in Cybersecurity daily, I fully agree with that assessment. However, “better” isn’t “perfect”. You still need vulnerability management, monitoring and response tools. And this is an area where Linux lags behind Windows, mostly because of the effort put into it. Something as basic as Endpoint Detection and Response (EDR) on Linux is still a foreign concept to a lot of folks. And yes, it’s absolutely necessary. I’ve personally worked cases involving compromise of Linux systems and Linux specific malware. And more than just XMRig clone #5374. It may be harder to move from user to root on Linux than moving to local admin on Windows (unless some jack-off decided that Linux doesn’t need updates and DirtyCow still worked); but, there is still a lot of damage which can be done without root.
Overall, I’d say that the Linux ecosystem is evolving. It’s already embedded in a lot of environments of all sizes, but it’s usually a smaller part of a much larger Windows network. Hopefully, if enough large organizations start pushing away from Windows and towards Linux, there will be more investment and standardization. And, we’ve already seen some of that. Red Hat is a common distro of choice for large organizations because of that higher level of centralized control and standardization. Of course, a lot of the tools engineers want to use aren’t on Red Hat, specifically because of that centralization (read: everything AI). So, it will be interesting to see where we land.
Do you mean steam stats? I don’t know of many others, and those are going to be enthusiasts of one form or another.
There are a lot of other stats around, like statcounter, cloudflare, wikimedia, and even the data pornhub publishes from time to time. All of them show linux growing, and they all have different audiences and maybe methodologies
Yep thats the main one I’ve seen posted around. I’m sure their data is accurate but I’m not convinced it’s representative of the wider consumer market
~~https://en.wikipedia.org/wiki/Science,_technology,_engineering,_and_mathematics~~
My bad, I misunderstood
I wasn’t questioning STEM.
Op mentioned consumer OS market share stats, and the only ones I know of are from Steam.
https://en.wikipedia.org/wiki/Usage_share_of_operating_systems
The article acknowledges the difficulty in gathering data, and moreover for their desktop numbers they reference StatCounter which itself only counts web-based usage on a subset of pages which are part of their network.
Ah that’s my mistake