Re: Re: Reminder End of Survey: Relevance of Configuration Systems in Free and Open Source Software

Hello Xen,

sorry for the late reply, I just wanted to bring your answer up on the team
meeting before answering to it.

Xen <list@xenhideout.nl> schrieb am Do., 14. Juli 2016 um 00:32 Uhr:

> Thanks for the reminder.
>
> I just think statistical answers do not speak to the real reasons people
> have for doing things.
>
> I gave up after a few of those questions ;-).


Yes we had quite some dropouts, but that was to be expected with this
rather technical and specific topic.

Gabriel Rauter schreef op 11-07-2016 12:03:
> > I would like to gently remind anyone interested that the survey will
> > be closed on 18.07.2016.
>
>
>
> You don't have external reasons for "integrating" a configuration
> system. Such as "to improve the user experience" or anything of the
> kind. You do so because of real reasons but they are not covered by
> generalized categories. They say the best investors don't do market
> research. They go by their hunches and their appraisal of the situation
> stemming from their observative powers and overall awareness.
>
> First you make sure you understand everything. Then you make a choice.
> You don't do so out of "reasons". You do so because you can see it is
> the best or best-ly available way to achieve a certain level of
> functionality, or robustness, or flexibility, or configurability, or
> whatever, given the constraints of the environment. It is often a
> compromise but you just forge ahead with the best options you have
> available.
>
> Maybe that sounds like the things you have asked, but in that case they
> are obvious and you don't need to ask them.
>
> Anyway, regards, and bye.
>
> Anyway, I had written this several days before. Sorry about that.

This answer is as valuable to us as would be filling out the survey.
e.g. We have a free text form at the end of the survey so people can write
more openly.
Thank you for taking the time to write your thoughts on this topic.
We will make sure to evaluate it as part of our collected data.

As for a personal reply on your take for choosing the integration of a
configuration system.
I think your described method is what most of us developers do. Get a good
overview of available
technologies and choose what seem the best suited one with the least
downsides.
But the survey is not just about that.
We, the Elektra Initiative, are currently a group of students and doing
research and working on configuration systems.
Currently this is mostly doing research and working on libelektra and its
integration with different systems.
There are several goals behind this survey. For one some students of our
team will use correlations found in the results
in their bachelor thesis. Additionally the results will influence in which
direction further research and work will be done
on libelektra itself. As anonymized data will be made available at a later
date under an open licence
(It will likely be a CC one we are still discussing about the specific
clauses we will use) also other projects working
on configuration systems could benefit from it.
In general our team also hopes to raise some awareness for the topic of
configuration systems in FLOSS in general.

best regards,
Gabriel