Bug#1041440: popularity-contest: Non Debian - non Deb packages should be able to be reported - packages missing from Debian
On Tue, 2023-07-18 at 16:20 -0500, Karl Schmidt wrote: > While popcon seems a good idea - it seems that data from repository > downloads would do much the same job. User related statistics should be opt-in, turning on logs on the repository mirrors would mean it would be impossible to opt-out. > statistics on non Debian [.deb packages] These are already present in the popcon data, see "unknown" on: https://popcon.debian.org/ > and even non Deb package software installed. There are so many different packaging ecosystems that this would be infeasible. They also often have their own telemetry submissions. > I would imagine a setting to identify locations of non Debian > executables so such data could be collected. I think this is a bit too invasive, these could be stored in /usr/local or home directories, or /srv or /opt or random other directories. > There is a pseudo-package bug - wnpp - that almost no one uses, This is widely used, there are hundreds of open RFPs. Some of them even result in someone doing the work to package things, #1041232 for eg. > I would think that statistics on what is missing from Debian would be > quite important. In practice it is quite irrelevant, because something missing from Debian that lots of people use doesn't mean any person has the time, skills and motivation to package it nor that it will be possible to package or even redistribute. Usually contributors have some form of personal interest in the things they package for Debian, rather than packaging popular things that they have no interest in. -- bye, pabs https://wiki.debian.org/PaulWise signature.asc Description: This is a digitally signed message part
Bug#1041440: popularity-contest: Non Debian - non Deb packages should be able to be reported - packages missing from Debian
On 7/18/23 11:07PM, Petter Reinholdtsen wrote: [Karl Schmidt] While popcon seems a good idea - it seems that data from repository downloads would do much the same job. Due to the distributed nature of the mirroring setup, there are no such data, so it can not be used like that. It Would be possible to fix that. Data just from one server would give a pretty good idea of the popularity. People that run popcon likewise are a subset of the real picture. A scrip could look at the download logs - make a count - you could then compare with popcon to see if the numbers match - I think they would. What would be even more important is gathering statistics on non Debian and even non Deb package software installed. This has been discussed for a while. You might find for example https://bugs.debian.org/632438 > illuminating. Interesting. For example - I know there are a lot of people still using komposer - nothing really replaces it. The number of people running appimage packages in place of the older debian versions would be interesting to know as well. The non-debian executable information seems more important than popcon to me. -- Karl Schmidt EMail k...@lrak.net 3209 West 9th Street Ph (785) 841-3089 Lawrence, KS 66049 Modernity is the denial of uncertainty; a false narrative. -kps
Bug#1041440: popularity-contest: Non Debian - non Deb packages should be able to be reported - packages missing from Debian
[Karl Schmidt] > While popcon seems a good idea - it seems that data from repository > downloads would do much the same job. Due to the distributed nature of the mirroring setup, there are no such data, so it can not be used like that. > What would be even more important is gathering statistics on non > Debian and even non Deb package software installed. This has been discussed for a while. You might find for example https://bugs.debian.org/632438 > illuminating. -- Happy hacking Petter Reinholdtsen
Bug#1041440: popularity-contest: Non Debian - non Deb packages should be able to be reported - packages missing from Debian
Package: popularity-contest Version: 1.76 Severity: wishlist While popcon seems a good idea - it seems that data from repository downloads would do much the same job. What would be even more important is gathering statistics on non Debian and even non Deb package software installed. I would imagine a setting to identify locations of non Debian executables so such data could be collected. There is a pseudo-package bug - wnpp - that almost no one uses, but I would think that statistics on what is missing from Debian would be quite important. -- System Information: Debian Release: 12.0 APT prefers stable-updates APT policy: (990, 'stable-updates'), (990, 'stable-security'), (990, 'stable'), (500, 'oldstable-updates') Architecture: amd64 (x86_64) Kernel: Linux 6.1.0-10-amd64 (SMP w/8 CPU threads; PREEMPT) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE not set Shell: /bin/sh linked to /usr/bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled Versions of packages popularity-contest depends on: ii debconf [debconf-2.0] 1.5.82 ii dpkg 1.21.22 Versions of packages popularity-contest recommends: ii cron [cron-daemon] 3.0pl1-162 ii exim4-daemon-light [mail-transport-agent] 4.96-15 ii gpg2.2.40-1.1 Versions of packages popularity-contest suggests: ii anacron 2.3-36 pn tor pn torsocks -- debconf information: * popularity-contest/participate: true popularity-contest/submiturls:
