Packaging:
- Package ships autostart desktop file, which should be dropped:
+ A lot of people don't do microblogging, and generally we do not want to
launch applications by default.
+ Gwibber is written in Python, which uses an excessive amount of CPU, and
would ruin boot speed.
- Package has heavy dependencies (python-egenix-mxdatetime and python-
egenix-mxtools) which use 1.3 MB .deb size alone (for a Python library
which provides some date/time utility functions???); This is excessive
bloat, and will make the CDs overflow yet again. At least the
documentation/examples should be removed from those two python modules
(which account for 2/3 of the package size).
Bugs: Package has ~ 500 bugs ATM, of which many are untriaged and looks serious
(lots and lots of crashes). Of course many are wishlist bugs, which is okay and
just shows that it is a fairly popular piece of software. But still, this is
three times as much as firefox.
I wonder whether it is really mature enough to go into an LTS yet.
How much of a goal is gwibber for lucid? Is it an option to keep it in
universe and install on demand, so that it doesn't fall under the LTS
support commitment? It looks like no single person could keep up with
bug triage, and at least fix all those crashes?
** Changed in: gwibber (Ubuntu)
Status: New => Incomplete
--
[MIR] gwibber
https://bugs.launchpad.net/bugs/523387
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
--
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
