Eric Holscher - Posted in 2018

One Percent for Open Source

2018-03-09T00:00:00+00:00

The open source ecosystem is the most valuable part of the software industry today. From the programming languages to the web frameworks, the operating system to the cryptography, all software companies today use open source.

Open source software is creating billions of dollars in value for companies and venture capitalists, but it has a funding problem for itself. There are not good ways for open source projects to capture the value that they create, while being available for free and providing the fertile ecosystem that companies are built on today.

There are many warning signs that open source is an ecosystem in danger. Many people doing this work are burning out, giving up, or just never getting involved in the first place. We need to improve the sustainability of open source to protect the valuable ecosystem that we have built.

We need a new funding model for open source.

Funding the Open Source Ecosystem

I’d like to propose an idea:

A program called 1% for open source: where companies agree to invest 1% of their funding round or yearly profits into the open source software that they use.

This is a win/win/win:

Companies invest in the tools that they use to ensure continued development and support. The companies can also use their support of open source to make hiring good developers easier.
Open source projects are able to fund project management and maintenance that is required to make them sustainable for the long term. The project then continues to be a force multiplier for the entire software industry.
Venture Capitalists invest in the open source ecosystem that makes companies much cheaper to build. It’s also really good marketing for the VC firms, to show they are investing in our common infrastructure.

Why is this a good solution?

The biggest reason is because all parties involved benefit from open source. Having a concrete point in time also makes it much easier to define when the money should be invested. It applies both to startups and established companies, and mixes payments in terms of lump sums and yearly payments, to make budgeting easier.

We have tried many different methods of sustainability in open source, but many of them are project specific and don’t address the ecosystem. Increasing investment in infrastructure across the entire industry is the only solution that will work in the long term. My focus on this program is to increase the total revenue for the entire open source ecosystem.

A lot of places have “Open Source Fridays” and “20% time”, but those programs are the first to be cut when a company has issues. They also mix incentives, where the work that gets done is more valuable to the company than the project. We need a program that invests real money in the infrastructure that we use, with the projects themselves choosing their own direction.

How should we implement it?

The company should invest 1% of the money into the open source projects that they use. There are many different ways this could be implemented. A few of my ideas are:

Companies create a list of the OSS projects they use, using tools like http://libraries.io/, and then give money to the 10 largest projects that have a way to accept money. This could be done during the license compliance phase of the investment process.
There could be a new non-profit foundation created, which gets all the donations and then has established criteria for how to reinvest the money as grants.
You could give the money to the existing foundations that focus on the area of open source that is most valuable to you

Making the execution of the investment standard is an important part of making this actually happen. There should be standard processes and criteria for where your money will go, so that it’s not adding additional overhead to companies who give back.

A simple example for a Python project could be:

You give 1% of your VC investment to the Python Software Foundation
You run a script that looks at all the projects in your requirements.txt that your project depends on
That list is cross-referenced with a list of the Python projects you use that need funding
The top 10 projects you use that need funding, sorted by lowest amount received that year, are each given 10% of your investment as a sustainability grant

Prior Art

I have a few sources of inspiration for this idea:

One Percent for the Planet - They ask established companies to give 1% of their revenue to help invest in the planet. This model for sustainability of an ecosystem was my inspiration here.
Stripe’s open source retreat - where they fund projects to work on specific features.
Mozilla’s open source support - where they fund awards for projects that they use internally
Coinbase’s open source fund - which is a monthly donation program
Segment’s open fellowship - which gives 3 month fellowships, similar to Stripe

Help out

I was inspired to write up my thoughts because I tweeted the idea, and it was well received:

We should start a movement called "1% for OSS", a program where VC's and startup companies agree to invest 1% of their funding round into the infrastructure that their companies rely on. #sustain #sustainoss
— Eric Holscher (@ericholscher) February 23, 2018

I don’t plan on actually building a new program for funding open source, so please contact me if this is something that you’d like to actually work on! This is a viable option for funding large amounts of open source, but it will require a lot of work. I’m happy to help, but it will really take motivated people to execute on this idea and make it happen.

I see the next steps being:

Figure out the exact process for determining who gets money from a project
Figure out the proper entity to use (an existing non-profit, or a new one?)
Find a willing VC or Company to do a test run
Iterate on what you learn, establishing standard processes for the next test run
Once you have a model that everyone is happy with, start expanding the program

Thanks

Thanks to everyone who gave me feedback on this post:

Samuel Wright
David Fischer
Russell Keith-Magee
Nadia Eghbal

Conference Mentorship

2018-02-13T00:00:00+00:00

I have had the opportunity to provide mentorship to folks who have organized conferences twice in 2017. Through this process I have realized the value in this practice, and I’d like to write this to promote others to do to same for first year conference organizers.

I think this is a practice that would be very valuable to build inside the global Pycon & Write the Docs community conferences. Having a mentor who has done it before will make life much easier for the people putting together their first conference.

How to mentor

The most recent experience I had was doing this as PyCascades. Looking back, I would say the role could be defined as this:

Provide a working example of major documents that a conference needs (budget, volunteer & organizer schedules, sponsorship prospectus, conference schedule)
Provide perspective when people had questions about how to do something, or worrying about small details that wouldn’t matter. Focus the group on the right problems.
Play part-time cheer leader. Reminding people that we were doing great at something really hard, much better than a first-year conference had any right to be doing.
No major organizational tasks would be assigned to me

The total time commitment, outside of attending the actual conference, was around 25 hours over the entire calendar year.

The value provided

I believe the value in this was on a number a levels. One of the biggest was really psychological, I believe it put the other organizers at ease knowing I had done this before.

The major value that I provided as I see it was:

Having a starting point to work from, instead of creating things from scratch
Stopping additional work from happening that didn’t need to happen, based on my previous experience
Making people feel more confident that we could actually produce the event, making the entire thing possible in the first place
Reducing the amount of time spent stressing out over things that didn’t matter in the end

Examples

I think having a few examples would be useful here to really drive home the value.

Financial Aid

PyCascades was a first year conference. We were really worried about being inclusive and bringing in a diverse crowd. It was really hard to figure out how much to budget for financial aid, because it was dependent on how much sponsorship we got.

This was a major source of stress for the organizers, but I tried to provide perspective:

We’re a first year conference. Doing any financial aid is amazing, so we can’t beat ourselves up over not having enough!

We ended up having a great financial aid program that allowed folks to attend, but giving perspective really helped people feel good about what we were doing.

Sponsorship

Small conferences are often dependent on sponsorship to be great. We budgeted tickets to cover the basics like the venue, but depended on sponsorship for the nicer thing we were able to provide.

After a couple months of trying to get sponsors, the other organizers were feeling unsure about the response. It felt like nobody was going to sponsor and everything would be terrible. I knew that this was just part of the process and was able to resure folks:

It takes months for organizers to process sponsorship applications. This isn’t a simple purchase, but if we keep up the good work we’re doing, sponsors will come on board.

Thanks to the tireless effort of the entire team, we ended up with over 15 sponsors for a first year conference. But it didn’t feel that way from the start, and I was to show that it would just take time.

You should be a mentor

If you have organized a conference before and have a good grasp on the operations, you should help mentor another conference. It’s one of the highest leverage ways you can spend your time, if your goal is to help expand a community in the real world.

Next steps

I’d also love to open source the basic documents that I provided to the organizers of PyCascades from my work on Write the Docs. I think having a shared repository of basic conference starting documents would go a long way to making this a scalable system of organizing conferences, and allowing mentors to learn from each other over time. It’s all the same value that we get from open sourcing code, but for building a community in real life!

The post I never published

2018-02-07T00:00:00+00:00

Looking back, the file was created on my 29th birthday:

Jul  9  2016 read-the-docs-sucession-planning.rst

I had failed. I burned out. (I’m fine now, don’t worry!)

I started writing a blog post that you’ve seen around the internet time and time again. Another creator of another open source project steps down, offers the project to anyone who will actually help maintain the thing. Maybe someone steps up, maybe not..

Is this the only way to get people to actually contribute to my project?
Is it blackmail?
How does using the nuclear option affect my career?
What will people think?
Am I really a failure?
I could have done more

The post

I actually forgot that I had written it, until looking back in my drafts for ideas of new posts to write.

This post is a first draft, written in July 2016, and was never finished.

I’m sharing it because I think it’s important to talk about these things, and mental health is not something we talk about enough. I’m doing well now, which makes it easier to talk about.

Read the Docs Succession Planning
=================================

I've been working on Read the Docs for almost 6 years now.
It's a large part of my professional identity,
the thing that I'm the most known for.
I've always thought I would be happy to work on it forever.
However,
the leaves change color,
the world spins,
and I get older.

At this point,
I have started to feel jaded and less inspired.
I see the problems easier than new solutions,
and don't have the vigor of youth and possibility.
I still firmly believe that the project can do more good in the world,
but I become less convinced that I'm the one to do it over the long term.

Sadly,
I've been failing at growing a community of contributors around the project.
I don't fully understand why that is,
but it's something that I've become more worried about over time.
I don't necessarily have the energy to onboard new people like I used to,
and I want to make sure the project is in good hands for the long term.

So,
this letter is a public recognition of this failing,
and hopefully will also spur action on the part of some kind soul.

Read the Docs needs more people who are willing to work on it.
We need people to do just about everything in the project:

* Translation management
* Operations and Infrastructure work
* Documentation updates and improvements
* Support and GitHub ticket triage
* Design and UX updates and improvements
* Development of new features and new ideas
* Promotion of the project with blogging and social media

Just about anything that someone could be needed is needed.
We've had a lack of interest in development on the project for a long time,
and as such things have languished.

Get Involved
------------

We have a well-documented process for getting involved in the tickets:

* http://docs.readthedocs.io/en/latest/contribute.html

We also have hundreds of possible projects that can be done from the operations and development side.

That’s all that I managed to write.

Now

Things are better now. I don’t recognize the person who wrote that post. They seem distant, a faint memory.

Read the Docs is sustainable. As of January 2018, we have a team of four people paid to work on the project. It feels like we might still nudge the world. I feel like I could keep working on this project for a long time.

I got really close to quitting the project though. It was really hard. I struggled for a long time. For every person who goes nuclear, there are 10 who have gotten close. I’m glad that things worked out okay for me, but not everyone is so lucky.

I hope we can take some lessons from Read the Docs and apply them more broadly. I hope the work others are doing in this area makes it better. There are a lot of problems that I still don’t know the answer to.

There’s still a lot more work to do.

Business questions I’m working on in 2018

2018-01-04T00:00:00+00:00

In November 2017 there was a lovely gathering of independent business folks in Portland, Oregon called DazzleCon. Leading in to that lovely event there was an introduction thread where everyone explained their business. It caused me to really think deeply about the problems facing my two fledging but promising communities, and I wanted to record my answers here publicly.

These are the big questions that I’m thinking about for the communities I help shepherd in 2018.

My email to the DazzleCon attendees

Hey there fellow Zebras!

My name is Eric Holscher, and it’s been heartwarming to read all the work that everyone here is doing. I feel a little out of place, but hopefully I can explain what I’m doing in similar terms.

Most of my work is focused on improving documentation in the software industry. My view is that software, and open source specifically, can be a force for equality across the world, and documentation is a fundamental tool to lower the barrier to entry.

My work involved two primary things:

Read the Docs, which is a documentation publishing platform provided for free to open source projects.
Write the Docs, which is a world-wide community of people who care about documentation (we call them documentarians :)

Since we’re here to talk about business, both of these projects are trying to address problems with business models:

Ethical Advertising

Read the Docs launched an effort that we call Ethical Advertising. The elevator pitch is “newspaper advertising for the internet” – we don’t track users and only target our advertising based on the content of the pages users are looking at.

This has been our solution to a long-term search for sustainability in the open source infrastructure, which is the larger narrative I’m happy to folks about, and I think it mirrors a lot of the existing funding struggles for public goods and services.

Non-exclusive membership models

Write the Docs is primarily funded through the operation of conferences, and we’re looking to expand into some kind of membership model. We have a strong history of inclusiveness, which includes offering all of our content and access to our community for free.

We’re hoping to create a membership model that isn’t based on access to exclusive content or resources, but instead of framed more as an investment in the ongoing maintenance of the community and production of resources. From what I’ve seen, this funding model hasn’t been very effective, but I’d be really curious to talk about this more with other folks who might have experience in it.

Looking forward to talking more with y’all next month!

Cheers, Eric