In the following article, we will see how we can provide data on the use of our .deb packages to the Debian team through popularity-contest. Obviously, the fact of providing the data of the packages that we use the most will be optional and totally anonymous.
WHAT CAN BE USEFUL FOR THE PACKAGES WE USE THE MOST?
Some of the utilities that Debian knows the most used packages are:
- Use the collected data to determine the package that will be installed by default in the Debian installation.
- To obtain information about the architectures that most users use.
- To fear an approximate idea of the totality of proprietary packages that users use.
- The totality of collected data is public. Therefore anyone can consult them to know how popular a specific program is. As an example, we can see that Firefox has been installed by 93,580 users who have participated in the program, while Chromium “alone” has been installed by 31,123 users.
- Get an approximate % of users who use Debian Sid, Debian Stable, etc.
In conclusion, with all the data collected Debian will do everything possible to improve its distribution.
WHAT DATA DOES THE DEBIAN POPULARITY-CONTEST PROGRAM COLLECT?
For each of the packages installed on our Debian team, it collects the following information:
- The number of users who have installed the package.
- People who use a certain package on a regular basis.
- People who install a specific package but do not use it.
- The number of people who have updated a specific package recently.
- Users who have installed a package but have not provided enough information for their data to be taken into account.
- The architecture of the packages that we install.
- The branch of Debian or the distributions derived from Debian that users use.
HOW TO PARTICIPATE IN THE DEBIAN POPULARITY-CONTEST PROGRAM AND INFORM DEBIAN OF THE USE WE DO OF THE DISTRIBUTION PACKAGES?
At the time of installing a Debian operating system, we are asked if we want to participate in the Debian popularity-contest program in order to provide data on the use we make of our packages. Once we meet with such a request, we will only have to answer if we want to participate in the program:
Note: If you look at the screenshot you will see that the default answer is no. We’ll see if the canonical people are as reasonable as Debian in this section.
In the event that we have installed the system and we do not remember our choice, we must act as follows:
Initially, we will check if we have the popularity-contest package installed on our computer. For this we will execute the following command in the terminal:
lateef @ debian: ~ $ dpkg -l popularity-contest dpkg-query: no package matching popularity-contest has been found.
Since the package can not be located, it means that it is not installed. As it is not installed it means that I am not giving the data of use of my packages to Debian. In the case that I want to provide information to the Debian team, I will have to install the popularity-contest package by executing the following command in the terminal:
sudo apt-get install popularity-contest
During the installation, we will be asked if we want to participate in the popularity-contest program. As we want to participate, select the answer Yes and press Enter.
Once the installation is complete, we will be providing information on how we use our distribution packages automatically and anonymously to the Debian team.
If one day we decide to abandon participation, we should only uninstall the popularity-contest package by executing the following command in the terminal:
sudo apt-get remove--purge popularity-contest
Another option to stop providing information without uninstalling popularity-contest would be to execute the following command in the terminal:
sudo dpkg-reconfigure popularity-contest
Once executed, we will be asked again if we want to participate in the survey about the use of packages. If we want to stop participating, we select the No option and press Enter.
HOW IS THE DATA SENT TO DEBIAN AND HOW FREQUENTLY?
The process of data collection is done as follows:
At the moment that the popularity-contest package is installed, a cronjob is configured that automatically, anonymously and weekly sends the use we make of our packages to Debian.
So in this simple way, we will participate in the popularity-contest program without having to do anything at all.
HOW TO CONSULT THE DATA COLLECTED BY Debian?
If we want to visualize the statistics compiled by Debian, we just have to visit the following URL: https://popcon.debian.org
Inside this web page, you will find detailed information about the use made by the users of the Debian package. If you want to get information about the use of a certain package, you just have to do the following:
- Select the branch in which you want to consult the information. In the example we are performing, we select the stable branch by clicking on the option Stable reports.
- Next, I write the name of the package(corebird) I want to consult in the search box and press the Go button.
Once the process is finished I get the following information about the corebird package:
If you think it is convenient, you can consult much more information. If for example, they want to consult the totality of packages ordered from more to fewer installations of the stable branch they should click on the following options:
and the result obtained will be as follows:
Note: If we deem it convenient, we can also download the data in a text file to processed a Subsequent with a spreadsheet.
If we continue browsing the URL you can get additional information such as:
- See statistics on the use of packages in a specific branch, such as main, contrib or non-free.
- See the architectures used by the popularity-constest program participants.
Obviously, the fact of providing our data to a third party is a matter of trust. If you trust this third person or entity, there would not be any problem. As in my case, I trust Debian, my opinion is that there is no problem involuntarily and anonymously assigning our data. And if in addition, the data collected serves to improve my preferred distribution then better.
However, in my case, I do not trust so much in companies as for example Canonical. It seems that from version 18.04 Canonical will implement mechanisms so that users who wish to provide certain data on the use and hardware of their equipment.
Unlike Debian, it seems that the data collection by Canonical will be more aggressive. Some of the data that will be collected will be the following:
- Information about the hardware that our equipment has, such as the resolution of our screen, the available RAM, etc.
- The location selected in the installation of Ubuntu, etc.
- The time it takes to install Ubuntu.
- If we download the updates in the same Ubuntu installation.
- Taste and version of Ubuntu that we use.
Now we just have to wait and see how Canonical implements the data collection. Hopefully, it will help them to optimize a distribution that just installed it has an excessive consumption of RAM.