Some suggestions for missing data points that come up often and would be interesting to monitor the trend for them:

1. Session type (X or Wayland).
2. Monitor refresh rate.
3. Adaptive sync usage.

Would be curios how many people use Wayland.
I am still on good old X.

You forgot another important one: vi vs emacs :P

Last edited by razing32 on 3 Dec 2019 at 9:20 pm UTC

Quoting: The_AquabatI think one interesting stat would be which package management are u using (RPM, DEB, SLACK or ARCH). You could deduce that from the distro but I'm curious if there is ppl using flatpak or snap out there.

I'd say that can be taken out into a separate question, like "Are you using any sandboxing, containerization or virtualization for gaming?", since Flatpak and etc. are not package managers, but sandboxing technologies.

Last edited by Shmerl on 3 Dec 2019 at 10:37 pm UTC

Currently X, 60Hz no adaptive sync. Using opensource amd hardware and drivers.

Session type (X or Wayland) was finally added (thanks Liam)!

Total number of users in the stats continues to decline. That's concerning.

Last edited by Shmerl on 2 Feb 2020 at 8:49 pm UTC

Quoting: ShmerlTotal number of users in the stats continues to decline. That's concerning.

People adding themselves to it, while not being a regular reader to come back to update it so they drop off - that's what it mainly is because of the data cut-off.

How many active users are there on average?

Oh, thanks for the pointer, I missed this stats section:

https://www.gamingonlinux.com/index.php?module=website_stats

It would be useful to have the number of active users there, in addition to total users.

Last edited by Shmerl on 2 Feb 2020 at 10:53 pm UTC

Wasn't there a store and number of game bought each month statistics in the past ? When did it disappear ?

Quoting: amataiWasn't there a store and number of game bought each month statistics in the past ? When did it disappear ?

That was a very long time ago, when it was an external survey being done once a month.

Why did it end ? That was very interesting data. I think that would be interesting to see how store decisions reflect into linux user adoption.

Though someone had brought it up a few months/weeks ago, but I didn't think it was that important:
What about including the Display language in the GOL user statistics?

Why?
I notice more and more often that new Linux users ask in chats/Forums stuff about Linux, SteamPlay, Wine etc., but when forwarded to other Guides/resources in English they point out that they asked in the chat precisely because their english isn't that good, so the Guide/resource e.g. of a Desktop Environment isn't that helpful for them if its in English only.

How should the question be asked?
IMHO just asking "What language do you speak" or "What display language do you use" wouldn't quite cut it.
Why, because the key point here isn't only what language is demanded but also, similar to the "does it run on Linux" question, is the translation to this language sufficient.
In addition since GOL is an English Blog its also most likely, strongly biased towards English speakers(both naive and foreign), which would be the greatest point against including the language.
So than if the question would still be implemented to the survey how should it be done?
1. What Display language do you prefer?(probably the Native language of that person)
2. What Display language do you actually use?
3. Are you able to find troubleshooting information for (Gaming on)Linux in your preferred language?(e.g. Yes/No/dont care)

So why does questions?
1. to simply show the demanded languages
2. to show if those languages are actually available/usable
3. Yes= I can find troubleshooting information in my preferred language
No= I cannot find troubleshooting information in my preferred language
dont care= since I speak english(thats why I am on this blog) I usally try both searches or maybe unconsciously english first, so I dont know about

Other notes/discussion points:
Display language would refer to the Distros display language while troubleshooting information would refer to all sort of things.
GOL is in english but I doubt that there are more than 70% native english speakers so the survey could still be useful, even if most non native speakers here obviously are capable of understanding english quite well.
What do you guys think? Should something similar to my proposed option be included? If so what should be adjusted/changed? Should such a part have more or less specific options than what I wrote?
And finally how should it be displayed on the statistics page?
Maybe ppl that use english as display language should be taken out in one of the displayed statistics similar to the "English only" Linux market share part?

Maybe it can help finding reasons why english(88.60%), Russian(2.87%) and German(2.19) are overrepresented in the Steam Hardware survey on Linux compared to all Platforms(English 35.02%, Simplified Chinese 26.27%, Russian 10.78%, Spanish - Spain 5.22%,...).

I don't know whether it would be feasible, but I'd love to have some adjustable cutoffs on the statistics trends graphs.

The ones with more than just a handful of items in them are crowded to the point of being unreadable, especially in the lower values.
The worst case has to be the chart of GPU models, but honestly, anything besides CPU/GPU vendor and driver ;)
I know one can select individual items by opening the legend, but again, with more than just a couple items, it's a bit tedious and error-prone, and not altogether obvious to begin with. Maybe a double-ended slider (right there on the vertical axis even?), or simply a pair of input fields for upper and lower bounds, could help?

Regarding the language suggestion:
As a data point, I both prefer English as my display language, and use English as my display language, be it OS or applications/games/what-have-you, although it is not my native language.

Every single time I have looked at localised versions, it just didn't feel right. Too many terms are loan words from English anyway, and more often than not, there are not insignificant portions that are untranslated.
I find that more distracting than having everything in my second language, even though "English" itself often means some funny dialect of Spanglish, Germish, Ruslish depending on the developer's background ;)
Still far less "immersion breaking", if you will. I have to admit though that I have always had an interest in languages beyond what school teaches, but besides a handful of words in a handful of other languages, I really only know enough of any of them to be dangerous at best (including English, probably) :D

Having language statistics for "China versus the rest of the world" might be interesting wrt. marketing, but only if the data can be collected automatically from a large population; I don't think including it here as a question would yield any valuable information, and at worst might degrade into some kind of popularity contest.

I am also not a native English speaker but still use English as the display language on my systems.

The reason for this is two-fold: 1. it helps with interoperability, since guides written in English match with what I am seeing and I occasionally produce Linux content and having my display language set to English helps with that, and 2. I am more fluent in computer-terminology in English than in Finnish. When I wrote my bachelor's thesis I had to invent terminology because field-specific terminology is either poor or non-existent. In English I feel I am more capable of expressing certain concepts and particularly certain specifics more accurately, because it is the lingua franca of computer science.

Quoting: ValckI don't know whether it would be feasible, but I'd love to have some adjustable cutoffs on the statistics trends graphs.

The ones with more than just a handful of items in them are crowded to the point of being unreadable, especially in the lower values.
The worst case has to be the chart of GPU models, but honestly, anything besides CPU/GPU vendor and driver ;)
I know one can select individual items by opening the legend, but again, with more than just a couple items, it's a bit tedious and error-prone, and not altogether obvious to begin with. Maybe a double-ended slider (right there on the vertical axis even?), or simply a pair of input fields for upper and lower bounds, could help?

Or maybe some kind of "zoom"(to a new chart or so) for the last few %?

Quoting: ValckRegarding the language suggestion:
As a data point, I both prefer English as my display language, and use English as my display language, be it OS or applications/games/what-have-you, although it is not my native language.

Every single time I have looked at localised versions, it just didn't feel right. Too many terms are loan words from English anyway, and more often than not, there are not insignificant portions that are untranslated.

Quoting: SamsaiI am also not a native English speaker but still use English as the display language on my systems.

The reason for this is two-fold: 1. it helps with interoperability, since guides written in English match with what I am seeing and I occasionally produce Linux content and having my display language set to English helps with that

Exaxtly this, I mean we all speak English but imagine someone who isn't so tech savvy and also not very familiar with English.
Trying to find a guide in how to run a Lutris script or how to use a Proton GE version can be a real down turner for most of this people.
No help in the respected language would be the same as a borked game on ProtonDB, and bad translation of Distros and other Basic tool probably a bad experience at best.

Having language statistics for "China versus the rest of the world" might be interesting wrt. marketing, but only if the data can be collected automatically from a large population; I don't think including it here as a question would yield any valuable information, and at worst might degrade into some kind of popularity contest.

Well kinda yes, but in that case a question like "Can/Could you comfortably use your preferred language" would be enough, Steams language statistics could account for the rest(which languages are over/under represented on Linux).
Also I have to revoke one of my points,
it definitely makes more sense to have such a measurement on an international platform, on a French platform for instance there would probably be >90% French speakers while a English Platform most likely has a more diverse audience.

However instead of yet another statistic it might also be worth having a discussion about this similar to the Steam Deck Talk with Liam, Linux4everyone, Nick from TheLinuxExperiment and gardener. Maybe with a small poll and some research about the topic beforehand
and some special guests or so
that would also be an idea🤔.

Last edited by Nickname on 28 Mar 2022 at 12:26 pm UTC