Steam's September Hardware Survey For Linux

[mendieta]

So what you are saying is that for people who launch Steam through more than one OS in a given month, which of those OS's they'll get polled under is random? Well, doesn't that mean that primarily Windows Steam users who occassionally launch Steam in Linux out of curiousity might get counted as Linux users by accident? Sounds to me like the polling method is fair in that regard. If there's not enough samples being taken, that's a different issue. However, the month to month consistency of the results increases confidence in the survey's accuracy.

No, I didn't mean that. I mean a real bias, in that users who use one OS as much as the other, get exclusively polled in one (Windows). Even worse, uses who almost exclusively use Linux and fire up Windows every now and then,only get polled in Windows. Please see the bug report quoted above by RussianNeuroMancer ...

[Tuxee]

Any 64-bit Steam client yet?
Any 64-bit Linux games on Steam yet?
Any open source games on Steam yet?

Why?
What for?
Not needed.

Besides, I've never seen this mysterious survey popup - and I have started my client daily since it got first installed about 10 months ago.

[RussianNeuroMancer]

Besides, I've never seen this mysterious survey popup - and I have started my client daily since it got first installed about 10 months ago.

Please add your comment here: https://github.com/ValveSoftware/ste...mment-25550027

[benmoran]

It IS easy to dismiss this as just the nature of random sampling, but it's almost certainly a bug. I run the Steam client on Linux almost every day since release on my desktop, and pretty often on my laptop as well. The survey has never showed up. I loaded Steam ONE TIME in Wine a month or two ago, and sure enough the survey popped up. Back when I used a Windows partion for gaming, I would sometimes get the survey a few times a year. It's been about a year since the Linux client was released, and nothing.


I know some people are convinced that it's just the nature of random sampling, but I don't think so anymore. If you remember, a few years ago Steam had a survey bug that was incorrectly reporting the number of Intel graphics users, so having bugs in the survey is not unheard of.

But that said, I really don't care about the survey numbers as long as new games keep being released. I'm sure developers care more about sales than survey results anyhow.

[dh04000]

Please add your comment here: https://github.com/ValveSoftware/ste...mment-25550027

Wrong one, see first page of thread for proper link.

[dh04000]

Besides, I've never seen this mysterious survey popup - and I have started my client daily since it got first installed about 10 months ago.

I turn steam on maybe every 3rd day and I've gotten asked to participate 3 times since steam for linux was first released.

[jophos]

I run windows and linux and I got polled on Linux this time. Its not baised, you get polled on whatever OS your on at the time you get selected to poll. I spend >90% of my time on linux, so it makes sense that it was very likely that I was on linux at the time of polling. Statistics.

Why do people call things biased if they don't agree with them? Its like dealing with the climate change deniers.

You hit the nail on the head. This is just the nature of random sampling. You may never get picked to do the survey, but a representative proportion of Linux users likely would have been. I have been asked to do the survey a couple of times while logged into the Linux client, but even if I hadn't been I would still have the same opinion.

I think that many people are suffering from confirmation bias. 1-2% sounds like a pretty reasonable figure to me.

[madjr]

It's not a real number, and it doesn't represent the importance of Linux. Why do we keep seeing it in the front page in Phoronix. Beyond me.

because it will be cool to see how Linux/steamOS gains more and more from now till it surpasses macOS and then we get as much or more games than mac users and from there Msft starts freaking out.

[smitty3268]

I think that many people are suffering from confirmation bias. 1-2% sounds like a pretty reasonable figure to me.

Yeah, and it just happens to line up with most of the browser based stats you find as well. And the same people will complain about how horrible those stats are too, because they don't want to believe in them.

[iwesp]

And the same people will complain about how horrible those stats are too, because they don't want to believe in them.

Considering the state of proprietary drivers and the limited selection of games, I'm actually quite happy that we even reached 1%.