Encouraging F-Droid Participation by Developers

[F-StopMember]

I am a proponent of F-Droid. I’ve noticed that some developers do not want to publish their apps on F-Droid, for a variety of reasons.

I think it may be best for F-Droid if the F-Droid team takes a look at the reasons why some developers do not publish on F-Droid. By doing so, perhaps changes can be implemented to encourage everyone to participate. Note that changes might include code/policy changes or may simply be an improved articulation of existing rationales.

For example, see:

https://github.com/M66B/NetGuard/issues/23
and
http://forum.xda-developers.com/showpost.php?p=63561767&postcount=48

[krtModerator]

I’d really love to see more app developers directly maintaining their apps. We indeed do listen to developers, but in the Android world there are unfortunately a lot of devs don’t reading their own licenses. If you are floss, you cannot permit us building (and signing) apps. Most of the time we just follow their requests, since it makes no sense to work against upstream. If someone wants that specific app, he most likely should start a fork.

But as I said, we liten to their complains. In regards to the signing thing: We see the problem and have implement ways to reproducible build apps and use the official binaries, when they match our builds. While this works, it requires both — upstream and us — to standardizie on a build process, which forces them to adopt fdroidserver. Most devs don’t care, because they don’t see a benefit of it.

About release times: We build once a day more or less. Problem is not building, but signing. Signing our apks requires human interaction for entering passwords. Reproducible builds wont require signing, if they match upstream releases.

There is more to talk about here, but most issues basically boil down to: we are lacking manpower: If we had per-app maintainers, those could spend more time on the app, our build and communicating/working directly with upstream.

[M66BMember]

The problem I have with fdroidserver is that it is too big and has too many dependencies, especially for a headless server, so maybe it is an idea to have a ‘lite’ version.

The following NEW packages will be installed:
  acl ca-certificates-java colord consolekit cpp cpp-4.7 cryptsetup-bin dbus dbus-x11 dconf-gsettings-backend dconf-service desktop-file-utils dmsetup dosfstools eject
  fdroidserver fuse gconf-service gconf2 gconf2-common gnome-mime-data gvfs gvfs-common gvfs-daemons gvfs-libs hdparm java-common javascript-common kmod libaacs0 libasound2
  libasyncns0 libatasmart4 libatk-wrapper-java libatk-wrapper-java-jni libatk1.0-0 libatk1.0-data libavahi-client3 libavahi-common-data libavahi-common3 libavahi-glib1
  libbluray1 libbonobo2-0 libbonobo2-common libcairo-gobject2 libcanberra0 libck-connector0 libcolord1 libcryptsetup4 libcups2 libdbus-1-3 libdbus-glib-1-2 libdconf0
  libdevmapper-event1.02.1 libdevmapper1.02.1 libdrm-intel1 libdrm-nouveau1a libdrm-radeon1 libdrm2 libexif12 libflac8 libfontenc1 libfuse2 libgconf-2-4 libgconf2-4 libgd2-xpm
  libgdk-pixbuf2.0-0 libgdk-pixbuf2.0-common libgdu0 libgif4 libgl1-mesa-dri libgl1-mesa-glx libglapi-mesa libgnome-keyring-common libgnome-keyring0 libgnome2-0
  libgnome2-common libgnomevfs2-0 libgnomevfs2-common libgnomevfs2-extra libgphoto2-2 libgphoto2-l10n libgphoto2-port0 libgtk-3-0 libgtk-3-bin libgtk-3-common libgtk2.0-0
  libgtk2.0-bin libgtk2.0-common libgudev-1.0-0 libgusb2 libice-dev libidl0 libieee1284-3 libjs-jquery libjson0 libkmod2 liblcms1 liblvm2app2.2 libmpc2 libmpfr4 libnspr4
  libnss3 libogg0 liborbit2 libpam-ck-connector libparted0debian1 libpci3 libpciaccess0 libpcsclite1 libpolkit-agent-1-0 libpolkit-backend-1-0 libpolkit-gobject-1-0
  libpthread-stubs0 libpthread-stubs0-dev libpulse0 libsane libsane-common libsane-extras libsane-extras-common libsctp1 libsgutils2-2 libsm-dev libsmbclient libsndfile1
  libsystemd-login0 libtalloc2 libtdb1 libudev0 libusb-1.0-0 libv4l-0 libv4lconvert0 libvorbis0a libvorbisenc2 libvorbisfile3 libwbclient0 libx11-dev libx11-doc libx11-xcb1
  libxau-dev libxaw7 libxcb-glx0 libxcb-shape0 libxcb1-dev libxcomposite1 libxcursor1 libxdamage1 libxdmcp-dev libxfixes3 libxi6 libxinerama1 libxmu6 libxrandr2 libxt-dev
  libxtst6 libxv1 libxxf86dga1 libxxf86vm1 lksctp-tools module-init-tools ntfs-3g openjdk-7-jdk openjdk-7-jre openjdk-7-jre-headless opensc pciutils policykit-1
  policykit-1-gnome powermgmt-base python-imaging python-libcloud python-magic python-simplejson sane-utils tzdata-java udev udisks usbutils wwwconfig-common x11-utils
  x11proto-core-dev x11proto-input-dev x11proto-kb-dev xorg-sgml-doctools xtrans-dev
0 upgraded, 185 newly installed, 1 to remove and 0 not upgraded.
Need to get 127 MB/136 MB of archives.
After this operation, 375 MB of additional disk space will be used.

• This reply was modified 1 year, 6 months ago by  M66B.

[M66BMember]

And please contact me at marcel(plus)netguard(at)faircode.eu for directly maintaining NetGuard.

[dudeofxMember]

well you can’t alleviate being tight fisted. Some developers like the prestige that of saying I do open source but at heart are still tight fisted with their work. I saw that with Nvidia a while back. I came to the forums cause I saw a lack of new apps on the fdroid app and I was thinking there was something wrong. If this is the reason blah, let those developers go.

The authors let go of their rights by releasing open source but they can still make life hard for the community. I saw that with Nvidia. Let them go. F-Droid is a good thing for the community. I think its their loss.

[M66BMember]

@dudeofx please read this to understand why deterministic builds are important: https://blog.torproject.org/blog/deterministic-builds-part-one-cyberwar-and-global-compromise

It is not about making anybodies life hard, it is about security and about support. Netguard is being developed and supported in my free time, but I am not going to support any non-deterministic build, because there can be any kind of modification.

[dudeofxMember]

@m66b well it should be your perogative. If you don’t want to support something for what-ever the reason you shouldn’t be strong-armed. I am just saying F-Droid is a good thing. and I am just here cause there was only one app on the whats new category. I thought something was wrong.

[Author]

Posts