You can’t gather these info from the forum. It’s OK to add the package, but not to enable its service by default without user consent.
4 Likes
Maybe I didn’t say it well enough, I meant the fields that the user can enter in their profile
But yes, disabling by default service would be some compromise
2 Likes
There is of course the limitation that there are only fields sufficient to cover one machine and may well not be kept up to date, or even properly populated in the first place.
1 Like
It should be opt-in, period.
The question now is, what do the users get if they opt-in? You need to give something back and make it collaborative. You can’t just take data and give nothing back.
One idea would be a tool that compares their data with the average Manjaro user. Something like User-Benchmarks. Users share their data on User-Benchmarks because then they can see how their system compares to others.
Be creative, make a nice looking tool with graphs n’ stuff that gets populated once the user enables data sharing. Find at least something to give back once the user opts-in as an incentive for sharing the data.
EDIT: I just ran inxi and thought, if something like inxi could be run in “collaborative mode” and show a comparison of my data against the average, that would be cool already. Every time the user runs inxi in collaborative mode, he shares his data and gets an up to date comparison against the average.
Is Manjaro Metrics not good enough for you? 
@romangg, the tooltips for the map are still appearing off-screen for the most southern locations unless one runs the browser window full-screen.
Maybe, if you can leverage that into making users share their data. That alone isn’t enough I’m afraid.
I gave an idea. It’s how people share their data willingly. People could just check their system against some loose statistics on the Internet, or… they could run the benchmark tool and get a personalized evaluation of their system.
It’s up to you guys to find something useful for the user that makes him click “Hell yeah I opt-in, this is cool”. A desktop widget, a weekly system health status, something. It can be little nice thing, and I’m confident, people will opt-in.
Well they do give us the distro for free, along with unlimited updates also for free. Asking for more, because the want to count every install, is in my opinion silly.
3 Likes
Um, that is the processed version of the data which the mdd tool at the subject of this thread uploads.
This. 
2 Likes
Now make it available to people who opt-in only and you’ve got your incentive. Bundle it inside some desktop widget or something where data is only shown when the user enables sharing.
Sorry. Linus Torvalds isn’t asking for my data, nor is Arch, from where Manjaro gets a bulk of the free updates they provide from.
Just being honest here. This is not an argument to ask for my data.
3 Likes
The Trends section has lots of pie charts for hardware data that can be filtered by distribution
hwinfo is available from Manjaro extra repository for gathering data; but is not included on Manjaro ISOs; and is not suggested on forum much; so not many users opt-in to uploading data
I use alsa-info.sh for in-depth information about hardware, drivers and ALSA, but alsa-info online data is not searchable like hwinfo.
I have asked many forum users to run sudo alsa-info.sh --upload and I do not recall anyone objecting to sharing information
An opt-out is not a bad thing as long as the user is properly informed and consents to share data. If I was asked to share data similar to agreeing to an update; and I could see data uploaded was benign; I would probably agree
“donor” is not totally inappropriate as sharing information would be useful for Manjaro Team,
but it is only a one-way thing
IMO if this was a “two-way street” and was useful for users it would get more acceptance
If user shares data with Manjaro Team and has a link to uploaded data they could also post the link to data in Troubleshooting topics
4 Likes
+1
for map : The tooltip position is offset because it doesn’t account for the page scrolling.
fixed in next or last version
1 Like
Maybe the geomap bubbles and legend on the site should show count instead of percent. Cause right now there are 3 bubbles >50% 
And i still raise an eyebrow on the graphic showing systems with 3 cores but 16 threads. Maybe loose the threads, the new intel cpus have no hyperthreading anyways, and amd is 1:2.
no
these are in fact 2 separate graphs.
not exists any link between cores and threads 
(we can clic on labels “cores” / “threads” for separate graphs)
they’re not all on the same machines
for example, read:
3 “pc” have 16 cores and 13 “pc” have 16 threads
3 “pc” have 12 cores and 3 “pc” have 2 threads
2 pie charts side by side would be more explicit…
or ?? if possible,
use a stacked bar chart (color threads : % in each bar) 
cpu count |o|
|x|
|x| |0|
|#| |-|
|#| |-|
6 (cores) 24 (cores)
color "#" % with 6 threads
color "x" % with 12 threads
color "0" % with 24 threads
color "-" % with 32 threads
``
1 Like
You are off track here.
You are fear mongering.
The data are clearly identified.
You are free to do what ever you want on your systems.
1 Like
If Linus or Arch asked for it, my answer would be the same. I don’t mind sharing basic info.
3 Likes
I know, but it just looks that way. Maybe 2 separate graphs would be better.
Neither Linus nor Arch is Manjaro. Arch doesn’t need to make a living from Arch Linux, and Linus makes his living from the Linux kernel, not a distribution. Quite different.
The fact is, I can understand business’ perspective while at the same time I understand users’ perspective, and in my humble opinion, mdd is a very good compromise. It doesn’t share anything personal, you can check and see for yourself.
And as I have it, they’ll (have to) ask for the information, not just take it, which is quite different in my opinion.
But, if you still don’t like it, maybe is GNU Linux-libre is more to your liking.
Edit:
Choice to turn it off again. Choice to keep using Manjaro or use something else. Choice.
Oh, and as a bonus you get a free operating system. One that only asks, not just takes, a few non-personal pieces of information.
4 Likes
As I run mdd, I get:
Welcome to MDD - The Manjaro Data Donor
Preparing data submission...
Error: "failed to connect: Host is down"
Traceback (most recent call last):
File "/usr/bin/mdd", line 843, in <module>
main()
File "/usr/bin/mdd", line 808, in main
data = get_device_data(args.telemetry)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/bin/mdd", line 753, in get_device_data
"audio": get_audio_info(),
^^^^^^^^^^^^^^^^
File "/usr/bin/mdd", line 530, in get_audio_info
"active": 'core.daemon = "true"' in pipew_out and not pulseaudio_active,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: argument of type 'NoneType' is not iterable
Looking at the metrics map “worldwide usage”, broken up in countries - what is the source for this?
The data I can see in mdd --dry run do not convey a country, do they?
"locale": {
"region": "de_DE.UTF-8",
"language": "de",
"timezone": "Europe/Berlin"
},
Probably the IP address the submission is from.
My country is (apparently) quite big, so that would mean nothing in identifying me, personally.
1 Like