[Themaintainers] Open Source Software: What makes maintenance something people do ?

[Atul Pokharel]

Hi Maintainers,
 I've been spending a lot of time (since about 2013)  thinking about why
people *stop* maintaining shared infrastructure (thanks for the shoutout
Andy).  Jan, I'd love to talk to you more about how your motivations
evolved. I'd also love to learn from any other maintainers about how their
motivations changed (It doesn't have to be software). If you'd be willing
to talk, please drop me a line. The gist of my approach is the observation
that the negative reasons - why people stop - seem to be more helpful for
understanding the moral reasons underlying maintenance than the positive
reasons - why people continue. These reasons can be different because
motivations evolve, people learn, maintenance becomes onerous etc.

In relation to some of the comments in this thread, I am fairly well versed
in the Ostromian approach to shared resources (Governing the Commons, 1990
and a ton of other works). This falls in the "law and economics approach"
that Christopher mentioned above.  It is a very common way to analyze free
and open source software, particularly when the objective is to come up
with metrics for characterizing the projects themselves or to compare
different cases. It is also, a very common way of studying natural resource
commons (like forests), physical infrastructure commons (like irrigation
systems), and digital infrastructure (like open source software).  Some
years ago, there was also a big push to try to derive a policy framework
for governing the "knowledge commons" (http://knowledge-commons.net/gkc/)
using this approach.

Theoretically, a major weakness of this framework is that it conflates
initial cooperation (required to build something) with sustained
cooperation (required to maintain something). This is largely because of
the underlying model of human beings that it uses  (whose motivations for
not doing something are assumed to be the exact opposite of the reasons for
doing something) but that is another story. There are other key
distinctions that these economic frameworks (largely) fail to make: that
between fairness and cooperation (it assumes that if people are
cooperating, they think it is fair), that between initital motivations and
subsequent motivations (assumes that everyone is isolated) and that between
natural resources and infrastructure (assumes that maintenance means
essentially the same thing in both cases).
And practically, very little of the evidence underlying these frameworks
follows the same cases over time. It is more likely to look at snapshots
across many resources at a point in time.

If you are into doing some research in this area, I am happy to help
however is most useful to you. I am also currently working on a project (to
submit for the Ford/Sloan grant) to study what fairness means in open
source projects, and whether we can come up with indicators of fairness to
throw in the bag of considerations for assessing the health of projects,
what can be done to help, etc.. It would be awesome to work together.

In my own studies so far of decades old farmer managed irrigation systems,
I found that people's motivations evolved so that moral motivations became
more salient (I followed the same cases over three decades). In particular,
a major reason why people stopped was that they came to perceive their
system as unfair.  I also found that reasons for stopping maintenance were
far more likely to concern moral elements while reasons for initiating
cooperation were some mix of personal pleasure, gain, promise, etc. Hence
the early comparison of reasons people stopped maintaining (
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2802653) . There are a
lot of questions here around what fairness means in the context of
maintenance, how it changes, whether user governed irrigation systems are
comparable to modern infrastructure  (yes, surprisingly so), and what
frameworks we should use to understand these moral motivations underlying
maintenance. I am currently working on a book that presents one
possibility  (working title: Fairness and Cooperation). It is encouraging
to see that the observations it is based on might not be completely wrong.

I am still really curious though how these lessons from many thousands of
years of physical infrastructure governance translate into our new digital
world and the other way around...

thanks,
-atul




On Tue, Aug 25, 2020 at 1:21 PM Bruce Caron <bruce at nmri.org> wrote:

> Prof. Neera Singh at U Toronto is working on the role of emotional
> commitment (affect) as a feature of commoning, <
> https://geography.utoronto.ca/profiles/6318/>.  I suspect her work on
> community forests in India might also apply to open-source software.
>
> bruce caron
>
> On Sun, Aug 23, 2020 at 12:24 PM jan <dittrich.c.jan at gmail.com> wrote:
>
>> Hello Maintainers,
>>
>> TL;DR: How are people socialized to continue to maintain (open source
>> software)?
>>
>> I wonder if you know any studies that analyze what makes developers
>> continue to maintain a piece of code, often without pay. A lot of open
>> source culture celebrates (libertarian) freedom, creating new solutions and
>> avoiding obligations (at least according to Nafus, 2011 [1] ).
>> Code/Software is provided "as is", so there is no written social or (or
>> even legal) contract to NOT leave a project, yet people seem to go through
>> a lot of pain to maintain old code or adapt old code to changing
>> infrastructure. I thus wonder how and why they continue to maintain in the
>> face that a lot of the openly celebrated activities are somewhere else.
>>
>> Jan
>>
>> [1]: ‘Patches don’t have gender’: What is not open in open source
>> software,  Dawn Nafus, 2011
>> http://journals.sagepub.com/doi/10.1177/1461444811422887
>> _______________________________________________
>> Themaintainers mailing list
>> Themaintainers at lists.stevens.edu
>> https://lists.stevens.edu/mailman/listinfo/themaintainers
>>
>
>
> --
> Bruce Caron, PhD
> Executive Director
> New Media Studio
> Santa Barbara, CA USA
> http://www.nmri.org
> http://cybersocialstructure.org/
> _______________________________________________
> Themaintainers mailing list
> Themaintainers at lists.stevens.edu
> https://lists.stevens.edu/mailman/listinfo/themaintainers
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.stevens.edu/pipermail/themaintainers/attachments/20200827/2640a239/attachment.html>