Re: [Savannah-users] what are the usefulness criteria for submitted code?

[john smith]

> John, we have made suggestions of improvements
> in your scripts that would easily make them
> closer to something that could be accepted.
> Note that.

I noted that. I noted that several times now, and I thought I could not
have been more clear about how little I appreciate the manner in which it
was done.

I was going to take some time to digest what assaf told me and start
working on a constructive policy proposal, but since you people keep
digging at me personally, let's start here and now.

Please take note of how what I say in this post will be pertinent to the
Savannah policy and relevant to every admin and user (read to the end),
while the posts I cite below are off-topic attempts to discuss the
internals of my code, which the critics themselves characterize as useless
to nearly everyone.

Thomas Harding's post was the least problematic one.

> You should split your project by tasks <snip>

This is an attempt to contribute to my project, apparently with the goal of
making it more acceptable in the eyes of Savannah admins, even though I've
made it clear by then that I absolutely refuse to meld my code in exchange
for hosting. I refuse to do so in principle, and not just because there are
just too many hosting options, including but not limited to my very own
iron on the backbone. I refuse to do so mainly because I believe that
validating this process will reliably lead to censorship and abuse on the
part of Savannah admins, if it hasn't already.  I have submitted my project
to Savannah under a false impression the rejection process is objective,
and once I was told by Karl Berry that it's not as a matter of policy, I
have rescinded my submission.

https://savannah.gnu.org/task/?14370#comment9

To be perfectly clear, I do not anticipate *using* Savannah hosting while
this policy is alive, even as I am working up to *contributing* to it.

I did not get offended by anything Thomas proposed, but in my replies to
him I've made it clear that I am the project leader, that I have no desire
to cede creative control, and that all good-faith contributions to my
project should be directed to my private email. I repeated the same request
in my reply to André's questions.

Now I hope you can understand my surprise when I read what Matthew Carter
wrote:

> [your scripts] make tons of assumptions about
> directories and binaries existing on a user's
> system, without proper safeguards / defensive
> programming around the different commands prior
> to running.

> I think if you spent all your time on those
> emails enhancing the scripts instead, they
> probably could have been in a state that was
> good enough to be accepted by now.

This is yet another attempt to contribute to my project, but this time it's
not just rude and off-topic, it also betrays Matthew's refusal to
acknowledge me as a project leader by ignoring my repeatedly stated
submission guidelines. Either that, or Matthew actually didn't care to read
what I wrote in this thread.

And just when I thought it could not get any worse, André, who I thought
was the most reasonable of these 3 (he did previously go private upon my
request), decided to chime in:

> we have made suggestions of improvements in your
> scripts that would easily make them closer to
> something that could be accepted.

<snip>

> I think these scripts, the way they are now,
> do not justify making a project from them

I understand it is your right, people, to use this mailing list for
criticizing my code, however irrelevant it may be in regard to the
questions I raised. It is also your right to discuss your would-be
contributions in any public forum, including this one. But you shouldn't be
surprised at my reaction when you blatantly disregard my polite requests
for using a proper channel, since by doing so you tacitly reject me as the
BDFL of my very own software project.

Now let's talk business; let me explain how all of the above is relevant to
the policy discussion.

In a way I am thankful for the outpouring of the irrelevant and rude
comments, because it clearly showcases what I have already suspected: the
standing rejection policy is eating your project from the inside like a
malignant fungus would. The 3 (non-admin) posters I mentioned above are
seemingly unable to recognize how out of place their comments and tone are,
and I think that's because as members of this community, they've gotten
used to treating submissions in a way only a software project should. They
think it's OK to wash the submitter's bones, criticize, and give advice no
one asked them for. In fact, they seem to think it's OK to insult a project
leader by willfully ignoring contribution guidelines.

Let this not be interpreted as a jab at the posters. I am sure they have
the best interests of Savannah in heart, and they actually believe they've
done me a favor. But I also believe they themselves have become victims of
the subjective rejection policy.

And they most definitely did nothing to help me. They seem to be oblivious
about my qualifications as a programmer, and I can assure them here and
now, I am fully capable of seeing defects in the code I submitted. I am
fully aware of every single issue they've raised, and then some. And if
they wonder why I submitted code if I believed it to be subpar, I'll say it
again: because back at the time I simply did not realize a hosting service
admin would exercise subjective judgement over ~120 KiB of nontrivial, at
least somewhat useful code.

They are also oblivious about my intentions with respect to the future
directions of my project, and how can they not be? They refused or not even
tried to talk to me in a respectful manner I believe I deserve as the BDFL.
Every single suggestion they've made was misguided by their incorrect
assumption that I want to take my project in a place where some Savannah
admin would regard it as "useful" for the purposes of hosting it. I have no
such intention. I have my own very specific vision for what constitutes
general usefulness, and what I want my software to do in general. This is
the vision I am going to implement, and I am going to do so on the schedule
I set for myself.

This is but one extremely negative result stemming from the consistent
application of the subjective rejection policy. I am too old and too
thoroughly validated in my programming skill to be shaken by these
comments, but the Savannah project is bound to turn away and alienate
coders this way, especially the young coders and the neophytes, who often
have great ideas, but lack the skill to produce a masterful implementation.
Telling them what to fix will NOT result in them fixing the code, unless
they are complete pushovers. No self-respecting project leader will bother
with being tutored by a git host admin ;) There are just too many options
for hosting, and too many options for getting tutored, and absolutely no
good reason, to make the latter a necessary condition for the former.
(Savannah has a reason for doing so, but none of its users do.)

When I have time, I will attempt to describe all the other things that are
wrong with this policy. I do not have any solutions at the moment, but I
will try to come up with those as well. I do not even know whether killing
this policy is practically possible, and assaf's post made me realize it
may be a hard nut to crack. But I want to be very clear that the comments I
took apart above are both symptomatic of a deep problem within the
community, and completely unnecessary, harmful even, to the Savannah
project as a whole.