The Trade-offs of Unpaid Free Software Labor

Wednesday 13 November 2013 by Bradley M. Kuhn

I read with interest Ashe Dryden's blog post entitled The Ethics of Unpaid Labor and the OSS Community0, and I agree with much of it. At least, I agree with Dryden much more than I agree with Hanson's blog post that inspired Dryden's, since Hanson's seems almost completely unaware of the distinctions between Free Software funding in non-profit and for-profit settings, and I think Dryden's criticism that Hanson's view is narrowed by “white-male in a wealthy country” privilege is quite accurate. I think Dryden does understand the distinctions of non-profit vs. for-profit Free Software development, and Dryden's has an excellent discussion on how wealthy and powerful individuals by default have more leisure time to enter the (likely fictional) Free Software development meritocracy via pure volunteer efforts.

However, I think two key points remain missing in the discussions so far on this topic. Specifically, (a) the issue of license design as it relates to non-monetary compensation of volunteer efforts and (b) developers' goals in using volunteer Free Software labor to bootstrap employment. The two issues don't interrelate that much, so I'll discuss them separately.

Copyleft Requirements as “Compensation” For Volunteer Contribution

I'm not surprised that this discussion about volunteer vs. paid labor is happening completely bereft of reference to the licenses of the software in question. With companies and even many individuals so rabidly anti-copyleft recently, I suspect that everyone in the discussion is assuming that the underlying license structure of these volunteer contributions is non-copyleft.

Strong copyleft's design, however, deals specifically with the problems inherent in uncompensated volunteer labor. By avoiding the possibility of proprietary derivatives, copyleft ensures that volunteer contributions do have, for lack of a better term, some strings attached: the requirement that even big and powerful companies that use the code treat the lowly volunteer contributor as a true equal.

Companies have resources that allows them to quickly capitalize on improvements to Free Software contributed by volunteers, and thus the volunteers are always at an economic disadvantage. Requiring that the companies share improvements with the community ensures that the volunteers' labor don't go entirely uncompensated: at the very least, the volunteer contributor has equal access to all improvements.

This phenomenon is in my opinion an argument for why there is less risk and more opportunity for contributors to copylefted codebases. Copyleft allows for some level of opportunity to the volunteer contributor that doesn't necessarily exist with non-copylefted codebases (i.e., the contributor is assured equal access to later improvements), and certainly doesn't exist with proprietary software.

Volunteer Contribution As Employment Terms-Setting

An orthogonal issue is this trend that employers use Free Software contribution as a hiring criterion. I've frankly found this trend disturbing for a wholly different reason than those raised in the current discussed. Namely, most employers who hire based on past Free Software contribution don't employ these developers to work on Free Software!

Free Software is, frankly, in a state of cooption. (Open Source itself, as a concept, is part of that cooption.) As another part of that cooption, teams of proprietary software (or non-released, secret software) developers use methodologies and workflows that were once unique to Free Software. Therefore, these employers want to know if job candidates know those workflows and methodologies so that the employer can pay the developer to stop using those techniques for the good of software freedom and instead use them for proprietary and/or secretive software development.

When I was in graduate school, one of the reasons I keenly wanted to be a core contributor to Free Software was not to just get paid for any software development, but specifically to gain employment writing software that would be Free Software. In those days, you picked a codebase you liked because you wanted to be employed to work on that upstream codebase. In fact, becoming a core contributor for a widely used copylefted codebase was once commonly a way to ensure you'd have your pick of jobs being paid to work on that codebase.

These days, most developers, even though they are required to use some Free Software as part of their jobs, usually are assigned work on some non-Free Software that interacts with that Free Software. Thus, the original meme, that began in the early 1990s, of volunteer for a Free Software codebase so you can later get paid to work on it, has recently morphed into volunteer to work on Free Software so you can get a job working on some proprietary software. That practice is a complete corruption and cooption of the Free Software culture.


All that said, I do agree with Dryden that we should do more funding at the entry-level of Free Software development, and the internships in particular, such as those through the OPW are, as Dryden writes, absolutely essential to solve the obvious problem of under-representation by those with limited leisure time for volunteer contribution. I think such funding is best when it's done as part of a non-profit rather than a for-profit settings, for reasons that would require yet another blog post to explain.


0Please note that I haven't seen any of the comments on Dryden's blog post or many of the comments that spawned it, because as near as I can tell, I can't use Disqus without installing proprietary software on my computer, through its proprietary Javascript. If someone can tell me how to read Disqus discussions without proprietary Javascript, I'd appreciate it.

Posted on Wednesday 13 November 2013 at 17:00 by Bradley M. Kuhn.

Comment on this post in this identi.ca conversation.



Creative Commons License This website and all documents on it are licensed under a Creative Commons Attribution-Share Alike 3.0 United States License .


#include <std/disclaimer.h>
use Standard::Disclaimer;
from standard import disclaimer
SELECT full_text FROM standard WHERE type = 'disclaimer';

Both previously and presently, I have been employed by and/or done work for various organizations that also have views on Free, Libre, and Open Source Software. As should be blatantly obvious, this is my website, not theirs, so please do not assume views and opinions here belong to any such organization. Since I do co-own ebb.org with my wife, it may not be so obvious that these aren't her views and opinions, either.

— bkuhn


ebb ® is a registered service mark of Bradley M. Kuhn.

Bradley M. Kuhn <bkuhn@ebb.org>