| Author |
Message |
itusid
|
|
Post subject: statistics
 Posted: 01.06.2011, 13:50
|
|
Joined: 2011-01-05
Posts: 9
Status: Offline
|
|
what about a system that checks how many people really use aptosid?
and maybe something like "test pilot" for firefox.
a tool that reports info like software installed, hardware, ecc...
il could help to choose which software include, settings, ...
it's just an idea.  |
|
|
| |
|
|
|
 |
|
slh
|
|
|
Post subject: RE: statistics
Posted: 01.06.2011, 13:56
|
|
Joined: 2010-08-25
Posts: 962
Status: Offline
|
|
| We are not going to ship spyware (paraphrased) with nagscreens to submit these reports. |
|
|
| |
|
|
|
|
|
DeepDayze
|
|
|
Post subject: RE: statistics
Posted: 01.06.2011, 14:02
|
|
Joined: 2010-09-11
Posts: 616
Location: USA
Status: Offline
|
|
| exactly slh, why do we need that anyway? |
|
|
| |
|
|
|
|
|
itusid
|
|
|
Post subject: RE: statistics
Posted: 01.06.2011, 14:14
|
|
Joined: 2011-01-05
Posts: 9
Status: Offline
|
|
I don't see it as a "spyware".
I think, for example, test pilot is useful to see how firefox is used, so it could be improved and adapted to users' needs.
And the same for aptosid.
Just an example.
If with the tool you'll see that 90% of users use firefox, it could be considered the idea to install it by default...
And obviously it could be "transparent", it would ask to users if they want to participate or not.
I repeat... it was just an idea!  |
|
|
| |
|
|
|
|
|
zbreaker
|
|
|
Post subject: RE: statistics
Posted: 01.06.2011, 15:53
|
|
Joined: 2010-09-13
Posts: 8
Location: new york
Status: Offline
|
|
Yes, it would indeed be interesting to know indeed. But as mentioned I think Aptosid is better off without such a "feature"
imo. |
_________________
apt-get upgrade-life
|
| |
|
|
|
|
|
dpt
|
|
|
Post subject: RE: statistics
Posted: 01.06.2011, 19:52
|
|
Joined: 2010-09-11
Posts: 281
Location: New Delhi
Status: Offline
|
|
Please, please, please. Let Linux be Linux. The way it is. Especially aptosid.
Let there be no automatic changes based on some auto feedback using the logic of someone
trying to finish a project for his degree.
As slh has said, no spyware. |
_________________
In a lunatic asylum, everyone thinks that he is the doctor.
|
| |
|
|
|
|
|
domicius
|
|
|
Post subject: RE: statistics
Posted: 01.06.2011, 19:53
|
|
Joined: 2010-09-13
Posts: 35
Location: Zagreb, Croatia
Status: Offline
|
|
| There's already http://popcon.debian.org/ but I'm not sure if it could be used to determine whether people use aptosid and even if it could, since such things must remain fully optional, you still couldn't tell how many people use aptosid. |
_________________
domicius
|
| |
|
|
|
|
|
slam
|
|
|
Post subject: RE: statistics
Posted: 02.06.2011, 08:12
|
|
Team Member
Joined: 1970-01-01
Posts: 607
Location: w3
Status: Offline
|
|
Several issues:
1) We already fight with space on our release images, so every addition would need very good reasons.
2) All tools collecting such data are potentionally opening doors to abuse of the data, and we do not want to be helpers for such abuse. If it is just about knowing some general data about Aptosid usage, we do have this data already from several sources (mirror usage, iso downloads, stats collected for sidux.com, distrowatch, etc.)
3) As already mentioned above, Debian is using popcorn, which already gives us more than enough info about popularity of applications. Add that popularity is not the main reason for us to include a piece of software.
Greetings,
Chris |
_________________
an operating system must operate
development is life
my Debian repo
|
| |
|
|
|
|
|
DonKult
|
|
|
Post subject: Re: RE: statistics
Posted: 02.06.2011, 12:51
|
|
Team Member
Joined: 2010-09-02
Posts: 485
Status: Offline
|
|
|
domicius wrote:
There's already http://popcon.debian.org/ but I'm not sure if it could be used to determine whether people use aptosid and even if it could, since such things must remain fully optional, you still couldn't tell how many people use aptosid.
While the popcon data COULD be used in that way, its not possible cause the data is not accessible in that way for very good reasons.
Lets imagine the raw data would be available - it would be pretty easy to generate a list of packages i have installed just by looking at the user who has most-recent APT version installed. I really don't want to tell anyone that i have pornview installed on my system -- and popcon data could even tell you that i use it regular…
Even if i would be not an APT developer, with a bit of data you could track me down - you just need to know a few "obscure" packages i have installed. I guess is ~5 packages to nail down any given aptosid user. (kernel, language, desktop environment, browser, another non-default-installed package -- you could get all that from a screenshot or a bugreport)
Thats a problem e.g. if you want to compromise my system. Just find a security hole in the software i have installed on my system - you know exactly which version i have installed.
You can also analyze how often i do upgrades. And when.
So popcon raw data is a pretty interesting thing for data miners as you can do a lot with it - you just need to make sure that everything stays anonymous if you want to be a good person.
P.S.: No, i don't have pornview installed (currently). Its the primary example in many topics (some of them: 'system safe to use for kids', Privacy concerns and unexpected search results: 'unattended' [think of upgrades or home automation] or 'movie') so we use it regular as an example in talks about these topics. And because people tend to give us a funny look if they hear it the first time - ten people in a small room talking about pornview  |
_________________
MfG. DonKult
"I never make stupid mistakes. Only very, very clever ones." ~ The Doctor
|
| |
|
|
|
|
|
dpt
|
|
|
Post subject: RE: Re: RE: statistics
Posted: 02.06.2011, 18:34
|
|
Joined: 2010-09-11
Posts: 281
Location: New Delhi
Status: Offline
|
|
A fair guess would be by the no. of registered users, especially those who have posted at least twice.
But I see 986 total registered users.
Users using aptosid on regular basis without asking any question (making any post) seems a bit unusual. But so does the small figure 986. |
_________________
In a lunatic asylum, everyone thinks that he is the doctor.
|
| |
|
|
|
|
|
slh
|
|
|
Post subject: RE: Re: RE: statistics
Posted: 02.06.2011, 19:04
|
|
Joined: 2010-08-25
Posts: 962
Status: Offline
|
|
| In my experience only a small number of users take the burden of registering at a forum. So while the number of registered forum users is certainly an indicator, I wouldn't expect more than 10-30% (this already assumes a vibrant, small audience - it would be much smaller for large commercial projects) to register (and anonymous apt repository access from unique IPs per day suggests, despite us employing load balancing over 5 primary mirrors - with log access to only one of them, the same). |
|
|
| |
|
|
|
|
|
ghstryder
|
|
|
Post subject: RE: Re: RE: statistics
Posted: 02.06.2011, 19:25
|
|
Joined: 2010-09-12
Posts: 97
Location: Detroit
Status: Offline
|
|
| Almost every time I am logged on, there are just a few registered users and a lot of guests. Right now, there are 4 registered and 20 guests, as an example. |
|
|
| |
|
|
|
|
|
itusid
|
|
|
Post subject: RE: Re: RE: statistics
Posted: 03.06.2011, 11:21
|
|
Joined: 2011-01-05
Posts: 9
Status: Offline
|
|
ok...!!
as I said at the beginning, it was just an idea.
I didn't think it could be as dangerous for users, I was seeing it just like an improoving tool...
no matter. |
|
|
| |
|
|
|
|
|
|
|
Post subject: RE: Re: RE: statistics
Posted: 03.06.2011, 14:10
|
|
Team Member
Joined: 2010-08-26
Posts: 170
|
|
| there are already too many big brothers watching you I don't think we should introduce/create a new one |
_________________
Why do strong arms fatigue themselves with frivolous dumbbells?
To dig a vineyard is worthier exercise for men.
Marcus Valerius Martialis, römischer Satiriker und Epigrammdichter
Será que o diabo perdeu a virgindade?
|
| |
|
|
|
|
|
domicius
|
|
|
Post subject: RE: Re: RE: statistics
Posted: 05.06.2011, 20:04
|
|
Joined: 2010-09-13
Posts: 35
Location: Zagreb, Croatia
Status: Offline
|
|
itusid, don't take it personally if somebody criticizes "your" idea. I'm sure many others had the same or similar idea, but didn't post about it here. I thank you for doing so because I learned something new out of it, thanks to great insights of aptosid users and developers.
After all, a lot of people like statistics for one reason or another and you can see aptosid developers already have a rough idea about usage of aptosid by monitoring some other data in a way that's more acceptable to all of us, I hope. So your idea wasn't even that off... |
_________________
domicius
|
| |
|
|
|
|
|
|
