On Mon, Jun 17, 2019 at 04:10:03AM +0300, antisocrates wrote:
> I am looking for a sponsor for my package "i3-gaps"
Thanks for your contribution! I looked into your package and have the
* Did you talk to the i3-wm maintainer, Michael Stapelberg, about his
opinion on this, and also about ways how to avoid the conflict
between the packages and instead enable it to co-exist?
* Your Recommends is a bit too "fat" IMHO. I would suggest keeping
all that stuff aligned with the i3-wm package, and leave everything
else to the i3 meta package. After upload, you can then ask the i3
maintainer to include your i3-gaps package as alternative
dependency in his i3 meta-package.
* Please add a paragraph about what i3 is to your package
description, not only the differences in i3-gaps.
* Your d/copyright lists i3wm upstream as source rather than the fork,
and also lists the original author as upstream contact. It should
point to the GitHub source repo and list Ingo Bürk as upstream
contact, While at it, please also look into whether Michael
Stapelberg is the only copyright holder. At least Ingo's changes
might be big enough to be separate copyrightable work. How did you
verify the information you put into d/copyright?
* I would prefer to have your packaging repository on
* Hmm, two almost exact copies of the same code in Debian in two
conflicting packages? i3-gaps is a widely accepted fork I suppose,
yes. What is Michael's opinion on that?
Please go through my remarks and comment each of them with what you
think about it, and whether you implemented it, will do so or
seeing the popularity of vanilla i3 and by extension i3-gaps (which
some people prefer aesthetically) in conjunction with the combined
userbase of debian and all of its offspring, it became apparent i3-gaps
should have been available sooner in debian.
one was to graph the users the users of the debian-family of
distributions and every other distribution family out there, it becomes
clear that a lot of people might be missing something they are used to
using elsewhere (most other big families of distributions have the fork
available as can be seen here: https://repology.org/project/i3-gaps/versions).
a lot of people have real trouble compiling this piece of software from
source (mostly new linux users of the debian-derivatives).
since i use debian from time to time, when my gentoo box is occupied
compiling for 12 hours straight, it occured to me i should package
i3-gaps for debian and therefore all its derivatives to give people easy
Certainly, a merge is the superior
solution and i am glad its actually being considered, but my
understanding is that code refactoring/cleaning could take a
considerable amount of time, and therefore the package should actually
be available as a temporary solution for all those who want to use it,
as it currently is everywhere else.
Again, i congratulate Michael and Ingo for their work.
> Certainly, a merge is the superior solution and i am glad its actually
> being considered, but my understanding is that code refactoring/cleaning
> could take a considerable amount of time, and therefore the package should
> actually be available as a temporary solution for all those who want to use
> it, as it currently is everywhere else.
I am hereby refraining from sponsoring your upload:
1. You did not respond to my remarks (yet)
2. Both upstreams said the fork is of minor code quality
3. Efforts are being made to fix that
4. Debian stable is not for temporary solutions
You might find someone else to sponsor your upload, so I leave this
bug open. Please keep in mind that in the case that someone uploads
your package, you are responsible to keep it updated, and probably
manually fix all issues yourself if upstream decides to work on
vanilla i3 instead, and are also responsible for tidying up once the
fork is merged into i3 itself if that happens ☺.