The share of Linux users on Steam is still holding steady
Another month goes by and Valve have released their Steam Hardware Survey. It shows a drop in Linux marketshare, but it’s not as bad as people claim.
I actually stopped talking about this for a while until recently as not much changes. When I see people linking to articles with a title like this:
«Steam Linux Usage Saw A Notable Decline For June 2017»
I can’t help but attempt to clear it up.
To be clear, there is no feud between GOL and Phoronix, but I do take issue with, to be blunt, clickbait headlines like that. A large amount of his reporting is generally quite accurate. However, headlines like that and the text of the article too saying it’s a «relatively large decline» just aren’t right.
Linux marketshare on Steam dropped by 0.09% for last month. That is not a notable decline, it’s so low it could be within the margin of error. We don’t know what their sample size is though, so the error margin is not clear. It might look like more than it is because our overall share is obviously a lot lower than where Windows is currently. This could also be from variations in the sampling each month.
The main thing: We’re talking sub-1% changes here. If there was a notable decrease in Linux gamers on Steam, the decrease in the overall percentage would be a lot sharper.
It can be a simple case of Windows growth in certain markets outpacing Linux growth. That doesn’t suddenly mean Linux has any kind of notable decline in use. Linux is likely growing too, just not as fast.
You can claim the Linux survey doesn’t come up often enough, but that’s just how sampling works.
You can argue booting into your Windows dual-boot partition suddenly makes it appear and claim it’s wrong. Of course it would pop up then though, that’s it seeing a change in your system. So you are likely to see a survey.
The numbers in the Valve survey are very close to what I constantly see from developers when talking in public (and more often in private) about their sales numbers.
We’re strong enough for hundreds of indie titles each month, Feral Interactive, Aspyr Media & Virtual Programming all still port some of the bigger titles to Linux. It’s going well all things considered.
Источник
SteamLUG — The Steam Linux User Group
Attention: We run 2 — 3 events per weekend which are scheduled in a batch towards the end of the month. For input on event scheduling, visit here, for complaints about notifications, complain to Valve and then comment here.
SteamLUG (Steam Linux User Group) is a community of users (old and new) and fans of GNU/Linux [en.wikipedia.org] — GNU based Free Software operating systems which use the Linux kernel, which we sometimes refer to as just «Linux».
We also cover Steam for Linux news (Valve news, Steam news, and native games), as well as host game servers and regular gaming events (we have a nice event countdown timer on the SteamLUG website).
We always aim to be as welcoming of newcomers as possible, so please feel free to join us! 🙂
SteamLUG (Группа пользователей Steam и Linux) — это сообщество пользователей (как новичков, так и опытных) и поклонников GNU/Linux — операционных систем, построенных на базе свободного программного обеспечения GNU и использующих ядро Linux, которые иногда называют просто «Linux».
Мы освещаем новости Steam для Linux (новости Valve, Steam и игр для нашей ОС), содержим игровые серверы и устраиваем регулярные игровые события (у нас есть отличный таймер обратного отсчёта для событий на сайте SteamLUG).
Мы всегда стремимся быть доброжелательными к новым людям, насколько это возможно. Пожалуйста, не стесняйтесь присоединяться к нам! 🙂
SteamLUG (Grupo da uzantoj Steam kaj Linukso) estas comunumo da uzantoj (komencantoj kaj speruloj) kaj adeptoj GNU/Linukso — operaciumoj konstruas en bazo de la libera programaro GNU kaj uzas kerno Linukso, kiuj iafoje oni nomiĝi simple «Linukso».
Ni prezentu novaĵojn pri la Steamo por Linukso (novaĵojn pri la Valve, Steamo kaj ludoj pro nia operaciumo), havas ludgastigo serviloj kaj establi regula ludaj eventoj (ni havas bonega renombrilo por eventoj, sur la retejo SteamLUG)
Ni ĉiam aspiras esti bonkoraj al nova homoj kiom tio eble, bonvolu ne ĝenu vin aliĝi kun ni!
SteamLUG (Steam Linux User Group) è una comunità di utenti (nuovi e di vecchia data) e fan di GNU/Linux — i sistemi operativi liberi basati sul software GNU e che usano il kernel Linux, che a volte viene semplicemente chiamato «Linux».
Inoltre, trattiamo le news riguardanti Steam per Linux (News di Valve, Steam e sui giochi nativi), e ospitiamo server di gioco e eventi regolari su di essi (Abbiamo il tempo rimanente per l’inizio di ogni evento nel sito di SteamLUG).
Cerchiamo di essere sempre accoglienti verso i nuovi arrivati, quindi sentiti libero di unirti a noi! 🙂
Источник
Linux users in steam
Steam for Linux
7. Linux users are do-it-yourselfers
6. Linux users are comfortable with the command line
5. Linux users are security conscious
4. Linux users want customization
3. Linux users are accustomed to diversity
2. Linux users expect software to cost nothing
1. Linux users are aware of licensing
Update: Some points have been contested, point two most of all.
I demand to see proof of 2. Although reading the article, Byfield is a little more specific than the headline about what we pay for.
When forced to use Windows at work, I always install BASH. Though I’ve never used Ubuntu’s BASH.
Byfield forgot to mention our tolerance, patience, and helpfulness in dealing with beta software and post-release bugs and our expectation that all problems will be fixed (we know, as do many Windows users, many Windows problems won’t be fixed or that fixes will create new problems)
I just made a Tomboy note Finally to repeat so many SALIENT POINTS
Why Windows Sucks
Describe your new note here.
7. Linux users are do-it-yourselfers
6. Linux users are comfortable with the command line
4. Linux users want customization
3. Linux users are accustomed to diversity
2. Linux users expect software to cost nothing
1. Linux users are aware of licensing
Well, I know that I, for one, am not too knowledgeable with licensing. Others are probably the same about this.
But I’m just nit-picking here. This article probably should have used the terms «Linux power-user» or «Linux enthusiast» instead of just «Linux user», as that seems to encompass casual users as well.
This whole article seems to be based off of someone’s experience from being in Linux forums and active communities, which are obviously going to be dominantly populated by the people that have more passion for the platform.
Источник
Linux users in steam
Linux users are less than 1% of Steam’s total, but linux sales for some games are a lot higher percentage than that (like up to 10%). Are we all rich or what? 😉
Also that ratio can vary *a lot* between countries!
And then comes the question:
What’s wrong with Linux gaming in China and Japan? Is the issue with linux in general or also specifically with gaming?
(warning: data in the article is for a single game, but Liam does mention weak linux sales in China as a common issue)
And then another related question:
Is there more sales breakdown data like this available? Like. actual statistical data?
Helium Rain has something similar
Indeed! And it hit 14% Linux sales, wow!
Also from a 10 seconds search:
That’s because the OS usage statistics on steam are based on the surveys, not hard data. This is flawed in many ways such as:
Bias
Whichever OS is most popular gets more chances for its users to get randomly prompted for the survey, creating huge bias.
Linux users are less likely to agree to the survey
Quite simply, linux users are more likely to be more privacy minded and ignore surveys.
In the end, Valve knows EXACTLY what OS people are using, they don’t need the survey for it. that’s just public info. The real data is actually valuable (why would they give their competitors that data?), which is why they won’t reveal it. But based on their dedication to linux support it’s likely nowhere near how low the survey reports linux adoption is.
Bias
Whichever OS is most popular gets more chances for its users to get randomly prompted for the survey, creating huge bias.
This part doesn’t make sense.
If you have a universe of 1 billion users with 1% linux and randomly select 1 thousand you’ll get around 10 linux respondents.
Obviously you might get 7 or 14 or something because the linux segment is so small in that sampling, so the percentage will vary a lot.
But it won’t be always smaller, it will just jump all over the place. In other words, it would be imprecise, not biased!
Also Valve may need to hide the secret sauce to some extent, but IMHO pretending there is such low percentage of linux users when it isn’t actually would harm their endgame more than help.
Their big existential menaces are Microsoft walling them out of Windows 10 and consoles, plus some new stores that so far don’t care about linux too much, so they need/benefit from game devs wanting to port or develop their games for linux (and gamers wanting to come to linux) to fortify their position now and stand a better chance in the future.
That’s because the OS usage statistics on steam are based on the surveys, not hard data. This is flawed in many ways such as:
Bias
Whichever OS is most popular gets more chances for its users to get randomly prompted for the survey, creating huge bias.
Linux users are less likely to agree to the survey
Quite simply, linux users are more likely to be more privacy minded and ignore surveys.
In the end, Valve knows EXACTLY what OS people are using, they don’t need the survey for it. that’s just public info. The real data is actually valuable (why would they give their competitors that data?), which is why they won’t reveal it. But based on their dedication to linux support it’s likely nowhere near how low the survey reports linux adoption is.
I think that there is a more simple answer:
Most of the multiple account users (usually users that have other accounts for card farming) are on Windows. So that generates a lot of noice in the stats.
I am not saying that we are the 10%, but a 2% wouldn’t surprise me.
Bias
Whichever OS is most popular gets more chances for its users to get randomly prompted for the survey, creating huge bias.
This part doesn’t make sense.
As a Windows user, I’ve never agreed to the survey. As a Linux user, I do, as I want to see a higher percentage.
Most new users don’t think like that. You’re talking about a niche interest of an already niche community.
If you have a universe of 1 billion users with 1% linux and randomly select 1 thousand you’ll get around 10 linux respondents.
That’s only assuming the survey is
1. true random
2. done constantly
3. Every user agrees to its prompt
Since none of those conditions are met with the hardware survey, it’s likely innacruate like you say. or potentially even biased based on some other factors.
Also Valve may need to hide the secret sauce to some extent, but IMHO pretending there is such low percentage of linux users when it isn’t actually would harm their endgame more than help.
They’re not pretending anything. They simply have valuable user usage data that they don’t have any reason to share to their competitors. Almost every game client out there doesn’t provide any linux support. Even places that sell linux games like GOG. their optional client is windows only. If Valve has data that suggests there’s more linux purchases to be had on the horizon, they have 0 incentive to let others in on that. As the saying goes. never interrupt an enemy making a mistake
Their big existential menaces are Microsoft walling them out of Windows 10 and consoles, plus some new stores that so far don’t care about linux too much, so they need/benefit from game devs wanting to port or develop their games for linux (and gamers wanting to come to linux) to fortify their position now and stand a better chance in the future.
Absolutely this is one of the core reasons for the linux support overall. But there’s a bit of chicken and egg situation here too. Do more people come to linux because it has support (and other things), or is there more support for linux because people are using it more? Are valve adapting to an emerging market or are they trying to forge one deliberately so they can keep from being overtaken by MC’s «one ring to rule them all» cloud computing strategy?
Источник
