I want to compensate for work and encourage volunteering, but there's a dilemma
There are few thoughts about paying work that I would like to hear thoughts about.
More or less the process in principle should be something like this: members submit what tasks should be done, there's an estimate how much funds would had to be used to pay for compensation for the tasks, which are then approved and prioritized democratically for funding.
Now here's an example of what it might look in practice:
I submit an idea for sending email notifications to users when their toots have been replied to. People approve of the idea, and want to implement it. Turns out it demands maintenence labour, and we estimate that 10 dollars a month is a good compensation for it.
Someone is willing to do it voluntarily instead. Surely we should allow members to contribute to the coop this way, and we allow it.
Someone won't do it entirely voluntarily, but would do it for 5 dollars instead of 10 dollars. Should we give that person the task instead?
Someone is willing to do some of it voluntarily, so the person receiving 10 dollar compensation needs to do less. Should we lower the compensation?
If we answer yes to both of those questions, it leads us in a situation where the system would look in practice something like this:
- People could submit ideas, approve or disapprove of them, and choose which approved ideas are prioritized on equal basis.
- The person who does it voluntarily or for least compensation from common funds gets to do it.
Now if you are suspicious of me writing of ”giving a task to someone who is willing to do it for the lowest compensation”, I am too, it does scratch in my ear too! But I've explained my reasoning why it seems like a conclusion to be drawn if we want to simultaneously encourage compensating labour and encouraging volunteerism.
Probably a wise thing to do is to look at things from a case-to-case basis, but it might help to have some guidelines in place. Would love to hear your thoughts about this.
Also, I don't know if email notifications would demand maintenance labour, so my example earlier might not be ideal, but I hope it makes my point clear enough.
Leo Sammallahti Wed 14 Feb 2018 8:19AM
That makes sense, thank you.
There I just wonder if giving the job and the compensation for one person, which certainly has practical advantages you describe, would be less ideal for us as opposed to a more "open" system in the sense that everyone is allowed to contribute and be compensated, with the funds used for this sort of compensation shared more widely instead of concentrated on one person.
But then again, while this might be ideal, it might not be practical, at least for now, for good reasons that you point out.
Erik Moeller Wed 14 Feb 2018 10:53AM
My sense is that if you want a broad base of participation, it's more important that the paths for volunteering are clear. Right now in the case of social.coop that's not really the case yet -- we don't have a clear process for on-boarding people into different volunteer roles. This page is an example of something we need to flesh out:
https://wiki.social.coop/Social.coop_Roles
Ideally I think we'd want to also have something like this:
https://fedoraproject.org/wiki/Join
Note that the page is structured according to the audience: what kind of person are you? The built in assumption is that there will always be stuff to do. I'm not saying everything about the Fedora project is great -- just that this kind of thing seems helpful.
I should note that it can be useful to have a lightweight process for reimbursing small expenses within defined categories ("train ticket to conference to talk about social.coop"). It's the open-ended bounty model that I find most problematic.
Money can help or hurt communities, depending on how it is applied. When larger sums of money are available, I personally like grant or fellowship models that reward initiatives from individuals and teams as a way to complement salaried work.
Michele Kipiel Wed 14 Feb 2018 12:59PM
Thanks for sharin the Fedora model. We can definitely learn a thing or two from that!
Matt Noyes Thu 15 Feb 2018 2:28AM
I really like the Fedora volunteer framework, an easy way for people to find a way to contribute and to coordinate the work.
Erik Moeller Fri 16 Feb 2018 1:31AM
Yeah, the projects that have been engaging with online volunteers at a large scale for a long time generally have lots of lessons to offer. When it comes to dealing with $$$, it's also worth checking out the Fedora expense reimbursement process, e.g., for North American ambassadors.
https://fedoraproject.org/wiki/FAMNA_Reimbursement_Guidelines
Note the two dimensions of the process: a regular meeting where people can get approval for certain expense categories (e.g., travel, hosting an event), and a request tracker that is used to keep electronic records.
So to the extent that social.coop ever wants to fund outreach activities, that could be a model worth emulating in a more lightweight (and perhaps more democratic) way.
Matthew Cropp Sat 24 Feb 2018 7:26PM
I've been thinking about this question, and one thing that came up for me was the question of how what and how much we should do "in house" vs. contracting out. In particular, for hosting/upgrades/etc., as we scale, would it make sense for social.coop to ultimately put out an RFP to the Tech Worker Co-op list Masto.host, etc., and then pick a bidder and decide on a contract for 6 months or a year at a time?
Nathan Schneider Sat 24 Feb 2018 11:17PM
Good question. I think this depends mostly on whether members would actually want to do this work. I would lean more toward having members do it, because it increases a sense of buy-in and participation. And it could lead to opportunities for paid work among members, which is an additional benefit of membership.
Erik Moeller · Wed 14 Feb 2018 7:44AM
My general bias is to minimize the number of transactions with human beings in which money is involved. Transactions with machines or largely abstracted services are a different matter. But when it comes to humans, every financial transaction creates overhead:
- the overhead of the person having to reserve time to do the work
- the overhead of thinking about the fairness of the compensation
- the overhead of managing "bids" and evaluating them
- the overhead of ensuring accurate review and accounting of the work
- the overhead of managing community expectations around volunteerism vs. paid work
- the overhead of becoming productive, in case you constantly assign work to different people.
For this reason, I am generally skeptical about excessive use of bounties and one-off contracts to get work done. Rather, I tend to favor a model where you pay a person a predictable amount, and handle the task allocation independently of the payment.
So, let's say you have $500/month for development and you have a very capable person guaranteeing you 10 hours/month of time. You then work out a process by which you prioritize how that time is used. You will probably prioritize basic administrative functions very highly, until those are done satisfactorily. Note that this also dissuades you from "shiny object syndrome" where you contract out small tasks without adequately addressing foundational work.
You don't need $500/month to start thinking this way--but it helps to at least cross a meaningful threshold where the money makes a real difference to the person contracted to do the work. Obviously, this process forces you to frontload the question of who you want to hire for the work, but there are well-established processes for this, and generally, capable volunteers will already have self-identified by the point you can offer reasonable compensation.