Funding and Coordinating Python Projects via Non-Profits

Bradley M. Kuhn

Friday 15 March 2013

Who Am I?

• President and Executive Director of Software Freedom Conservancy.

• A 501(c)(3) not-for-profit charity dedicated to promoting, advancing, defending, and developing Free, Libre and Open Source Software.

• I used to be software developer.

I Doth Protest Too Much.

• I really used to be a software developer.
  • Last project I worked on actively was a an online poker system in Python, using Twisted.

• The reason I care about any of this is I want developers to have the freedom to write Free Software.

• What does that mean today?

• How do we go about it?

Hacking Organizations

• I focus now on structures and centers for non-profit Free Software development.

• Like software development, difficulties are in the details, not the general ideas.

• My job: solve these details to help Free Software projects.

How Projects Start?

• Projects start roughly one of two ways:
  • company throws it over the wall.
  • an individual or group of volunteers who love it.
• “Throw it over the wall” has limited interest to/interaction with 501(c)(3) non-profits.
  • 501(c)(6)’s are a better home (e.g., OpenStack).
  • Reasons are probably off-topic.
• Conservancy is focused on helping the …
  • … ragtag band of volunteers who want the project to succeed.

Measure of Success?

• I’d be fool if I argued we don’t need for-profit company support for Free Software.

• We do.

• The question is about who is in the drivers’ seat in Free Software projects.

The Hidden Free Software Funding System

• Shortly put: “hackers find a way”

• Developers who care about projects find a way to code give back.
  • … or they quit and find a better place to work.
  • HP’s 20% time idea (adopted most famously by Google) capitalizes on this to attract talent.

• But it’s harder than it looks.

It’s the duty of all Free Software developers to steal as much time as they can from their employers for software freedom.

A Portrait of the Free Software Artist…

• Free Software developers face constant cognitive dissonance.

• Their patrons want the code bent to their will …
  • … and it’s tough not to oblige.

• This is our community’s version of the tragedy of the commons.

Throughout history, if you want to know who controls the world, look to who funds the artists of their time.

Charities as a Counterbalance

• I have the pleasure of very little cognitive dissonance.

• I serve one master and one master only:
  • the public good.

• I want developers to limit that cognitive dissonance, too.

How This Relates To Fundraising?

• Free Software won’t exist without funding from companies.

• But that’s not the only place funding should come from.

• Non-profits can be a hub of funding.

Practicing What I’m Preaching

• Right now, Conservancy has:
  • active contracts with seven contributors to …
  • … Mercurial, PyPy, and twisted projects.

• They’re paid by the hour to work on Free Software.

• Their “boss” is the community of users:
  • they must blog monthly on their work.
  • & send me a public commit log before I release payment on their invoice.

• This is a great way to get Free Software ought to be developed.

Where’s the Money come from?

• Example: PyPy

• Wait, the majority are from for profit companies?

• Didn’t I just get through saying we should avoid that?

• Well, it’s complicated.

The IRS is Awesome!

• For international audience: the IRS is USA’s “Mr. Tax Man”

• But, they’re also in charge of charities filings.

• 501(c)(3) charities are effectively a gov’t grant, in two ways:
  • the orgs pay no taxes.
  • donors get a tax deduction.

• IRS is the regulatory body.

IRS’ Public Support Test

• No company (nor group thereof) can control a 501(c)(3) charity.
  • for obvious “tax evasion? reasons.

• A charity can’t legally promise a company to do its bidding.

• Donations from companies must be spent in the public good.
  • under penalty of loosing charity status.

• It’s The best darn corporation-control firewall we have for Free Software!

Why Don’t We Do More?

• Well, even Conservancy’s largest projects …
  • … (and PyPy is one of the largest) …
  • have <= $100k/year throughput.

• Conservancy’s entire revenue (30-some projects!) is only $1.7mil.

• These numbers don’t change the world (alone).

• The provide that counterbalance I talked about.

Why Don’t We Do More?

• This is the only fundraising pitch.

• But I’d love to see more blue in this graph …
  • … & every other Conservancy’s project graph.

Why Don’t We Do More?

• We need a culture …
  • … where people give a little cash from their pockets …
  • … back to the projects they love.
  • … so Free Software doesn’t live on the teat of for-profit largess.
• And, frankly, more companies …
  • … should be giving back cash too …
  • … rather than just staff time.

Why Don’t We Do More?

• PyCon tradeshow floor last year was enlightening …
  • … I’d never before seen a tradeshow floor focused almost exclusively on recruiting.
• It’s clear good Python hackers have low unemployment.
  • supply & demand will leave them well-paid.
• Non-profits don’t pay that much …
  • … Conservancy has lost contractors to places like:
  • … Facebook, consulting companies, & even trade associations.
• Developers have a choice to make, too:
  • What matters more in your life …
  • … who your master is or how much you make?

Where Do We Go From Here?

• Growth in non-profits never works like VC-funded companies.

• Adding an employee or two a year is actually really fast growth.

• That’s the path Conservancy is on.

• I’d love to see other non-profits funding development …
  • … despite the political complexity.

More Info / Talk License

• URLs / Social Networking / Email:
  • Conservancy: sfconservancy.org & @conservancy
  • Me: faif.us, ebb.org/bkuhn & @bkuhn (identi.ca only)
  • Slides at: ebb.org/bkuhn/talks & gitorious.org/bkuhn/talks (source)