Dear Savannah hackers,
My name is Ivan Zaigralin, and my email is address@hidden I am
using this gmail
account because messages sent to this list from my own domain seem to
disappear into
a black hole.
I am currently in the process of submitting a project to the non-GNU part
of Savannah. I seem
to have hit an unexpected barrier: unexpected to me, but may be that's just
because my
expectations were out of line with reality, so I hope you can help me to
resolve this issue.
What I submitted was ~ 120 KiB of bash code + licensing information. To my
surprise, my
submission is not being accepted, and the reasons stated I will simply
quote:
"These are person-specific scripts"
"It doesn't seem to me that they could be generally useful."
"Yes, this is my opinion that doesn't coincide with yours."
"All this makes sense for personal scripts, but not for general use. They
are just not written with
such use in mind."
"I don't think there are real objective criteria for things like e.g.
simplistic package. We have to
use our judgement."
I want to draw your attention specifically to the fact that the reviewer is
using nothing but his
subjective judgment in order to decide whether my submission is "generally
useful". I also want
to make it absolutely clear, I have no complaints about this particular
reviewer, and nothing in
this post should be interpreted as a criticism of that person or any of his
actions so far.
The reason I find this surprising is this: FSF endorses Savannah as a
"hosting service":
"There are many services that will host your project's source code"
"Savannah is a community project, providing code hosting for your free
software project"
This endorsement is explicit in claiming that Savannah will host *my*
project, which I
understand as me preserving the creative control over the code I submit.
To contrast, the GNU project does and should make subjective calls as to
what constitutes
useful GNU software, just as the KDE project members make subjective calls
as to what
constitutes contributions useful to KDE. This makes sense because these are
software
projects, and when I submit code to them from the outside, it is implied
that they have the
creative control (or at least a greater share of it), and will make
subjective calls in line with their
unique and subjective vision of what their project should be and how it
should get there. Most
such projects also have very detailed descriptions of their subjective
visions; for example, KDE
is defined as "advanced graphical desktop, a wide variety of applications
for communication,
work, education and entertainment and a platform to easily build new
applications upon", and
much much much more, which really narrows down the scope of the project,
and makes it
perfectly clear that only the code implementing that vision will be
accepted. There is also
absolutely no surprise when senior members of the KDE team, who share the
creative control
over their project, reject code based on their personal and subjective
notions of quality and/or
usability.
So I was taken aback, to be frank, when I was told by the reviewer that my
project is not
accepted based on nothing but personal and subjective criteria having to do
with general
usefulness. After a lengthy inquiry, I still cannot locate any official
Savannah description of any
usefulness tests applied to submissions. I was fully expecting objective
criteria (besides
licensing), such minimal & maximal size in bytes, but I cannot find any
listed anywhere.
Indeed, I cannot even find any official subjective criteria, which would
make sense if Savannah
was in fact a software project. So it looks to me like my submission is
being held up based on
a personal subjective usefulness test which was applied to my project only,
effectively singling
it out. So with the information I have now, the only way to interpret what
is happening is that
Savannah is de facto a software project, whereas Savannah hackers assume a
share of creative control right from the start, from the moment of
submission.
Just like any community project, Savannah is fully entitled to make the
rules, but as an FSF
member I see an issue with endorsing Savannah as a "hosting service",
unless it actually is a
hosting service in a manner I described above, which brings me to my
questions for the
Savannah community:
Does or does not the Savannah project demand, allow, or abide by
filtering/censoring/rejecting
projects based solely on subjective opinions of its members (Savannah
hackers)? If yes, what
is the goal for such practice? If no, does the Savannah project expressly
forbid such practices
internally?
Thanks for your time :)
References:
FSF endorsement: https://www.fsf.org/blogs/community/savannah
My Savannah submission: https://savannah.gnu.org/task/?14370