The Ethics of Unpaid Labor and the OSS Community (2013)
ashedryden.com
ashedryden.com
- I lose ~ 2-3 hours a day every day on commute alone.
- I lose an additional 1-2 hours daily on getting my daughter ready, taking her to school, pickup from grandparents, and other daily needs of my daughter.
- My income is SIGNIFICANTLY lessened by the expenses of my daughter
If I was not a father, I would have the following:
- I could live 15 minutes away from work by foot, thus saving me between 1.5 - 2.5 hours a day.
- Living close to work requires most likely paying double my current housing cost, and having half my living space.
- Living close to work eliminates my need for a car, so some savings there.
- I will have 2 spare hours for extra sleep or misc. activity a day
- I will also have 2-3 extra hours a day not taking care of my daughter (homework, Daddy-Child time, planning meetups with friends, planning activities).
- No need to pay for extra food (time + money), no need to have homecooked food (I can survive on "ramen", child cannot), no need for gymnastics or other physical / social / mental development activity programs (lots of money saved).
Add it all up I am at ~ 4-6 hour disadvantage to the average 23-27 year old single male, with a significant lifestyle and monetary hit. If I was just to use 2 hours a day that the more free have to program an OSS project, and 2-4 hours on sleep, I would be better rested, more productive, and would contribute between 10 to 14 hours a week to OSS all while keeping my current lifestyle.
P.S. on top of that I work at a job who doesn't count ass-in-chair time, lets me work 9-5 (not 9-6), and is low stress. So already I am privileged over others in my exact position without this cozy 9-5 job.
P.S.S. on top of that as a divorced man, I also need to find time to date, so you can't argue that I too have time savings over a single man's need to date. Which mind you is really hard when you have a child.
P.S.S.S. Some people have a job that revolves around and/or encourages OSS contributions. So they get to contribute, all while getting paid for it in their standard job. This puts then in an even further advantage because they are now unfairly elevated above the average worker.
Yep, and people who are disabled (or act as a caretaker) are in a similar boat.
Ironically, you are probably a better bet for a lot of employers than that average young single male because you value stability. You're not gonna jump ship because you want a foosball table or expect to have lunch catered every day.
Unfortunately, the market for labor is notoriously unequal and illiquid. The desire to save via a large workforce of what you're calling "averge young single male"s often looks much better on a balance sheet than a few well-paid older folks. There's lots of different kinds of capital, but SEC filings definitely favor "objective" measures like number of desks filled.
So to say that I must drive myself crazy just to meet your need for me to have OSS projects on my resume, I rather work somewhere else.
Side note: I work at a dream job, getting paid below market value, and won't trade it in because... how many people work in a job where everyone is underpaid and wants to be in the office every day because it is just so amazing?
Regarding the main point I think it's a non issue. Companies are only interested in hiring OSS contributors to get a feel for their work. There are many other ways to do this. For example since a lot of my work is visual I maintain a demo reel. For others it may be articles you've written or even Stackoverflow contributions. It is not hard to get a job without OSS if you have some way to communicate your past performance.
Also the author says she knows "many" women who have had trouble contributing to OSS based on how they were treated. GitHub should make it easy to provide examples of this right?
I know women have a tough time in certain tech environments, but I'm really surprised to hear of repos where this is happening because perpetrators would have to do it publicly.
Please provide examples of the largest OSS community where this is happening and where offenders are not called out or reprimanded. I'll be glad head over there and back you up.
- Twitter - Facebook - IRC - Twitch - Reddit - Real world (some wear hoods)
Some are more anonymous, some less.
http://notepad-plus-plus.org/news/response-and-apology-4-sex...
> Also the author says she knows "many" women who have had trouble contributing to OSS based on how they were treated. GitHub should make it easy to provide examples of this right?
Now you're being deliberately obtuse. Not everyone uses github, and not all communication happens through github.
It's also important to apply Hanlon's Razor [2], "Never attribute to malice that which is adequately explained by stupidity.". Sometimes that stupidity (or ignorance) is the person rejecting your contributions, but it could also be your own stupidity (or ignorance). I have found that most people who try to contribute to open-source and fail have never read esr's classic "How to Ask Questions the Smart Way" [3]
If you eliminate all identifier's of gender by using non-gendered handles and initials for your first name in your git config, you can safely look for other explanations for your difficulties contributing to open-source besides your gender. It's called applying the scientific method. If you hypothesis is "I'm having trouble contributing to open-source because of my gender", you should obfuscate your gender in your contributions to control for that variable. Once you eliminate it as a possible confounding factor, you're free to come up with alternative hypotheses for your difficulties.
Personally, I don't have my name, gender, or location on my github profile. It's a community for contributing code, so all such metadata is irrelevant. I'm genuinely curious what percentage of github users cannot be determined to be male or female (which is a figure the author did not cite).
[0] http://en.wikipedia.org/wiki/Fundamental_attribution_error
[1] http://en.wikipedia.org/wiki/Occam%27s_razor
Community building is very, very hard work and it involves a lot more than linking to classics like ESR's smart questions. In fact, in my experience, those responses tend to put off newcomers with backgrounds that are different than programmers like me (and presumably you) who have been tinkering in this community since the early 90s.
YMMV, of course, but I think personalities are incredibly important and the most valuable thing you can do in building big communities is making sure that as projects grow they gain stewards, shepherds, or whatever metaphor you prefer that can welcome as well as gatekeep.
In fact, in my experience, those responses tend to put off newcomers with
backgrounds that are different than programmers like me (and presumably you)
who have been tinkering in this community since the early 90s.
Meredith Patterson's "When Nerds Collide" best describes how I'd like to see our community grow:
https://medium.com/@maradydd/when-nerds-collide-31895b01e68c
If esr's smart questions is off-putting, then you're probably bringing values into the community that I don't want to see become common, especially the value that you've made a solid attempt to help yourself before you asked others to help you. If you can't or won't do that, then you'll never ever succeed in this community and this profession/hobby, because constantly asking questions and answer them yourself is a fundamental skill.
Linking to an essay like esr's might be off-putting to women, no-binary and transgender developers, but it's also off-putting to cisgender developers. It's off-putting to any one who has grown accustomed to letting others do the work for them and who haven't developed a thick enough skin to accept a FAQ as an indirect response to their request for help. Lots of things about our community are off-putting to people regardless of gender identity or orientation. I myself am pansexual, identify as agender, but come across as cisgender (mainly because I choose my clothing based on comfort, fit and simplicity (I pretty much have all the same clothes and only need to make a determination on the color, since varying color is sufficient to eliminate the judgement that I wear the same clothes everyday)).
If the aspects of our community that Meredith describes in her essay is off-putting to some, that's a good thing in my opinion because I don't want a culture in tech that is so accommodating that it evolves into including so much of the rest of society that it develops into having a notion of intersectionality that no longer accommodates weirdos. I have on more than one occasion seen very capable engineers rejected as a candidate because one or more people interpreted their aspie behaviors and mannerisms as "creepy". For example, being aware and accommodating of the mind-blindness off others is as important as being aware and accommodating of the gender and gender identities of others.
My personal experience has been that identity politics drives people apart. It used to be that when I came across someone of a different gender or gender identity as my own, I paid it no attention. What mattered was that we all had engineering and programming things to nerd out over. Now I don't like going to engineering conferences anymore because the environment of gender politics is palpable. Instead I've directed more energy and time to IRC and Github where people are focused on more interesting things like the qualities of someone's contributions (code and ideas) and not their identity. If there is no culture of constructivism and proofs, it's probably not a forum I have interest in. Most conferences no longer represent a culture of constructivism and proofs. Only conferences like StrangeLoop, RacketCon, ErlangFactory, etc. interest me these days. Since they don't represent language communities that are also an active job market (ruby, python, javascript, go, php, java), they don't suffer nearly as much from identity politics overshadowing the computer science.
That said: a lot of us have constructed our identities through exploration online, so the notion that IRC and GitHub are not expressive of personal identity sounds a little weird to me. It's very, very hard to actually separate identity from the developer, especially when it comes to activities like bug triage. I don't have a good answer, but I think there's an assumption that what we can measure necessarily is what is important, when what we measure in terms of programmers is often what's easy to measure.
We were able to resolve that via some twitter messages and some time to let both of us calm down, but at the time I was ready to stop using redis, much less helping folks via bug reports and patches.
EDIT: Downvotes? I guess it's socially unacceptable to not want to contribute to OSS because someone else will just make money off your work.
I'm honestly not sure what you wanted. You contributed code to something, and people used it. Some of those happened to make money on code that included yours (I'm sure the impact yours had on their bottom line is debatable). Did you expect financial compensation? Did you just hope for whatever reason that nobody would make any money off any code that you wrote? If you didn't want people to "avoid" your software, you wanted them to use it, right?
OSS is freely given, sometimes by people who are paid to do exactly that. The term "unpaid labor" has a specific connotation akin to slavery or indentured servitude. The two are not comparable.
Complaining that someone somewhere makes money (at best, indirectly) from an OSS contribution is the same as complaining that Good Will resold my donation at a profit.
It has nothing to do with whatever political axe you have to grind.
Is this a legal term of art or common knowledge definition? It was unknown to me so I've tried some googling and am turning up primarily volunteer work and unpaid internships under that title. I know wikipedia isn't the greatest source, but over there "unpaid work" is a class of things including both slavery and working at a family business or coop.
It sounds a bit like you're buying a round of beers for everyone in the bar, and then you're upset people didn't compensate you for it.
I occasionally write stuff for my own use that might be of use to others. There's no downside to me if I release it, and the occasional upside.
Sometimes I need to apply a fix to some OSS project. Contributing helps me avoid having to maintain a fork.
Sometimes I write something at work that may be useful for others and convince my employers to let me open source it, and as a result I am also able to reuse the same software at home or in future jobs.
None of these have deprived me of income. Some of the work I was paid for. Some of it has indirectly led to income.
I find your expression of resentment curious. Their use of your software did not deprive you of anything other than in your own mind.
If your code is popular, why not offer support to, or seek out sponsorship from, the companies that are using your modules/applications? A revenue-focused startup is probably a bad resource to tap for this. But getting a few companies to spread out a few hundred dollars each from their engineering/marketing budgets isn't completely unlikely.
Otherwise, you can always recite the age-old mantra to cleansing open source entitlement resentment: "Patches welcome."
Come to think of it, it's not even clear you could effectively write such a license for software (the analogous artistic license is generally about performance/broadcast, which doesn't have an equivalent in software). Once the user has received the code from the licensor, legally the situation is out of the licensor's hands _except for the issue of redistribution_, which is what the GPL restricts.
There I didn't use the word "exploited".. are you happy now?
> I did not want to contribute to OSS anymore because watching other people make money off of my unpaid work made me feel undervalued and resentful.
Again, what did you think was going to happen to work you freely made available for people to use as they saw fit. They used it as they saw fit.
You are complaining that people did what you expressly allowed them to do. It's probably best that you stopped contributing, you clearly didn't understand what you were doing.
Community-developed software kinda runs on a stone soup model -- everyone brings something to the pot, everyone gets something back OUT of the pot, everyone gets something that is better than the sum of the parts contributed to it. This can be a good thing. It doesn't stop you from being miffed if you figure out that there's a restaurant in town that's coming, dropping off a bit of mirepoix, hauling off a cauldron's worth and selling it the next town over.
And in the macro sense, there are a lot of companies who have figured out how to use the communities that have developed around open source and free software as a low cost to free source of labor. And important infrastructure products like OpenSSL end up being used EVERYWHERE but starved for resources because everyone can just freeride off the work of a few people until something like Heartbleed comes along and exposes the costs of it. And there's probably at least a few products that are in the same boat as OpenSSL and we just won't know what those are until the next Heartbleed. And it makes it really, really hard to solve these problems when we can't even talk about them openly without people being personally attacked for discussing it.
That's a nice idea, and many people do operate that way, but in no way is it required, enforced or the norm.
Most people use OSS without ever contributing back. They find what solves a problem they have and they use it. They can do this because the explicit license allows them to do it. You might have had the intention of 'share and share alike' but the explicit permissions say otherwise.
> It doesn't stop you from being miffed if you figure out that there's a restaurant in town that's coming, dropping off a bit of mirepoix, hauling off a cauldron's worth and selling it the next town over.
It makes your anger over it occurring irrational.
Here's why you might consider contributing to OSS even though others use it to make money:
If you have contributed to a successful project it's a major leg up in getting hired by someone using that project
If you use that product in your own work, then you are paying that benefit forward and can expect some (additional) returns in the future
If you learn something new/interesting that is outside of your normal work-a-day stuff
Seriously, companies typically give zero shits about which software license you personally prefer, they just want to be able to look over what you have accomplished.
Now consider the original motivation behind the GPL, which is to ensure software freedom [1]. Being open-source is merely a necessary prerequisite for software freedom. The GPL is much less idealistic than permissive licenses, and more practical about ensuring its goals. RMS had no illusions about the motivations of people and companies; the GPL's absolute requirement for open-ness is an acknowledgement that there will be licensees who are more than happy to use FOSS software against the spirit in which it was intended.
One side effect of the GPL's demands for open-ness is that when FOSS usage reaches a critical mass, companies will have no choice but to employ programmers that work on free software. These programmers will then contribute back as necessary, creating a self-sustaining FOSS ecosystem. This is already happening with Linux.
In a perfect world full of truly free/libre software, we probably wouldn't have articles like this complaining about "unpaid OSS labor" and GitHub resume padding. The playing ground will be evened out, and there will be proportionately fewer developers contributing because of "love and a lot of free time." In that sense, developers who contribute to FOSS under non-copyleft licenses are doing a disservice to the rest of us who don't have that luxury.
[1] The phrase 'software freedom' is used very specifically to refer to software that satisfies the four freedoms: https://www.gnu.org/philosophy/free-sw.html
What you contribute and the quality of what you contribute defines the respect and influence you receive as a result (in a given community, of course). I don't see how that can be tied into "normal" social statuses as this author is trying to imply.
Thoughts?
https://en.wikipedia.org/wiki/The_Rise_of_the_Meritocracy
Thinking that it's a good idea is like what happened with "waterfall" development. Someone described all the problems with what big companies were doing, and then someone else picked up that description of all the problems and said "ooh! all the big companies are doing this! this must be a great way to do things!" and then everyone perpetuated that disaster for the next two decades.
The fact that "hacker" culture has uncritically picked up on "meritocracy" and doesn't realize it's a snarky joke about unfairness says a lot about the unbelievably sheltered and privileged position that most people reading this site enjoy, and our collective lack of social skills. By "social skills" I don't just mean, like, how to behave at parties, but also how to manage groups and how to be aware of the perspectives of other people.
Not to mention the fact that if your open source community isn't measuring merit on an objective, defined-in-advance, linear scale, with measures in place, then you are not even practicing the (unfair, broken) system of "meritocracy", you're just doing "rule by cognitive bias" ("biasocracy"? surely nobody is going to take that term and think it's a worthy goal to aspire to).
You can trust me, because my community _does_ practice objective, quantitative meritocracy, and therefore I am provably the best and most deserving person in this conversation: https://twistedmatrix.com/highscores/
A similar dynamic happens in the tech world. The people determining 'merit' have that same built-in bias against females, and will judge their work as lower than it should be. The bias need not be conscious, and in most cases the men think that they are being fair and honest, but it's there.
Even more pernicious than that though - you say "the men think they are being fair and honest", but research repeatedly shows that women are also less fair to other women. This is important to keep in mind because it reinforces the fact that most of the biases that work against objectively determining merit are part of the fabric of our society. They are biases that you will tend to hold even if they hurt you personally.
Many men bristle when they are confronted with this sort of research and say "I'm not a sexist" or "I'm not a racist". That's the wrong way to think about it. We live in a deeply unfair society, and unless we constantly, mindfully, consciously work against that unfairness all the time, we will fall victim to it ourselves and perpetuate it.
Except it doesn't. Racism, sexism, personal arrogance, etc. all influence the way that people treat each other. People who feel they are not being treated with respect may never make it to the point of contributing anything, or they may decide it's not worth the emotional burden of dealing with assholes.
Are there many people, regardless of gender or race, who earn less than $35k/year and are in a position to contribute to OSS?
FWIW, I contributed to an open-source project and was then hired to work on it full-time, and it has been a great experience.
In short -- life happens but then the jobs dry up and good talent is over looked or lost.
And, yes, it does happen to women more than men.
If you are hiring for a senior development position and you have a filter you can employ that will cut out 90% of mediocre or bad programmers (or perhaps more importantly make it much easier to identify them), but will also cut out 40% of good and great programmers, it's not a foregone conclusion that you shouldn't employe that filter. If you still get enough resumes to make a decision, you've just decreased the odds that you get a shit hire by $MATH percent.
If social justice is something you care about you probably don't want to do this. If building good software is something you care about, you probably do. I think most here would agree that a random developer with a large body of public OSS contributions is probably "better" than a random developer without. I say that as someone who has pushed to a public GitHub repository probably four times in the past two years.
Affirmative action type programs are often discussed/implemented as bi-modal.
Okay I can pee standing up, but other than than?
Also whenever these kinds of arguments are made Asians are always counted as white, even though more Asians work in the tech sector than should by chance.
In this context the "privilege" is your normal. It's not some extra perk you consciously take, it's that you don't get some shit that happens to other people.
If your house is robbed you can call the police and not worry that they'll arrest or shoot you instead.
You can do your job with little fear that your boss will want to sleep with you and impede your career if you reject his advances.
You can use public toilets without causing uproar and/or breaking laws.
And you can be blissfully unaware that such things are real problems, because it either never happens to you, or happens rarely and on a smaller scale than for some marginalized people.
People asking you to mind your privilege don't want you to feel guilty. They want you to understand and believe that they have to deal with things that don't happen to you.
No I just have to fear that I say the wrong thing at a conference and lose my job.
>People asking you to mind your privilege don't want you to feel guilty
That is a lie.