Steam Linux Usage In May Creeps Up To 0.57%

[phoronix]

Phoronix: Steam Linux Usage In May Creeps Up To 0.57%

Following the Steam Survey corrections beginning with the April numbers that came following slightly inaccurate numbers due to cyber-cafe reporting and at a time when there has been large swings in users due to popular games like PUBG, the numbers for May 2018 have barely budged on the Linux front...

http://www.phoronix.com/scan.php?page=news_item&px=Steam-Linux-May-2018-0.57

[some_canuck]

still waiting for my survey, eight plus months later...

[Chewi]

I have been playing the shit out of Rise of the Tomb Raider but I always use Big Picture Mode so as usual, no survey.

[ptyerman]

There was no problem shoving the survey in my face in Windows, but I've still not seen it while using Linux. One must ask if Valve is purposely trying to screw up the Linux results!

[geearf]

still waiting for my survey, eight plus months later...

It's a survey not everyone is supposed to get it...

[ptyerman]

For a survey to be anywhere near accurate, everyone using the product should get it. Anything else is a shambles and a controlled result!

[Luke_Wolf]

There was no problem shoving the survey in my face in Windows, but I've still not seen it while using Linux. One must ask if Valve is purposely trying to screw up the Linux results!

Unless there is actual evidence to the contrary I'm going to say no it's just incompetence and poor sampling on their part, leading to a large amount of variance across the board not just with the Linux results.

For a survey to be anywhere near accurate, everyone using the product should get it. Anything else is a shambles and a controlled result!

Yes and no, It is true that anything short of a full census is not The Truth(TM). However, with a large enough sample size, and a correct sampling methodology you can get a close approximation of The Truth(TM) that is effectively indistinguishable from it. The problem is sampling correctly is difficult because it's extraordinarily easy to bias the results which is very often taken advantage of by less than scrupulous individuals.

Valve does not have a reason to engage in such malice, however they are for lack of better terms an extraordinarily lazy company and their sampling methodology is a secret. So it's not unlikely that someone put something into place that whoever was in charge of writing the survey thought was clever but is an unintentionally biased methodology and it was never fixed or improved upon because it's hard and nobody cares enough in the company to try to improve it.

[Weasel]

I wonder if they can "detect" Wine and count it as Linux because if not, I'm willing to bet their Linux statistics are half, if not less, than they should be.

[xiando]

still waiting for my survey, eight plus months later...

Seriously, what is the point of such a silly survey? Put a simple [x] Anonymously submit hardware statistics in a configuration section (Privacy? or similar) and now Valve gets a way broader data-set to play with. It's not that hard to do.

Around %0.5 seems accurate enough, though, if 2% of desktop computer users use GNU/Linux then 1/4th of them using Steam is a pretty good result for Valve.

All we need now is the right major GNU/Linux exclusive AAA title and the nations will accept the GNU World Order.

[Termy]

I wonder if they can "detect" Wine and count it as Linux because if not, I'm willing to bet their Linux statistics are half, if not less, than they should be.

i think its pretty save to say that almost all winXP can be assumed as wine...