Hacker News Reader

I've been working on the same libre project for more than 25 years, and making a living from it for about 15, so I have my own perspective on this.

The biggest issue that I see is that even for things that are in some respects "finished", grants on the order of $5k do not change the maintainance picture very much at all. If there's a sudden crisis with critical infrastructure, people will step. But that's precisely what we want to move away from, and to do that the funding needs to be living-wage level, not single-issue grants.

It is awesome when those grants happen, and specific new features or compatibility are worked on. But the sustainability question is really not about that kind of work, for the most part. Somebody needs to actually be the guy in Nebraska and they need to consider that their role. Possibly it is just one role among a few, but it needs to be bigger than a one-and-done $5k-sized role.

The question is really how to redirect the streams of revenue that currently flow toward capital so that the people who work on OSS can do this as a living, not a part time calling. I don't see grants as a significant part of that.

A grant can be for a lot more than $5,000. It can be for as much as the grant-making org has and wants to spend. Grants can be given on an ongoing basis as well.

Now the Open Source Endowment is a very small organization that starts with ~$5k microgrants. It is not enough for a living but still should help maintainers not only financially but also by allocating attention.

As it grows bigger, the grant size will also grow. One can help with this by donating and bringing in new donors!

I think my point is that grants may not be the way to get what is desired. Most people need predictable, long term income. You can get stuff done with grants, no doubt. But that's not the question - the question is can you build long term sustainable maintainance mechanisms for OSS. I hope you're right and I'm wrong.

Software has its own lifecycle, and the funding should not stick indefinitely to a specific project. Meanwhile, our grant format might evolve into some type of limited tenured positions for maintainers, which support the most critical yet risky projects. But this target scope should be dynamic and adapt to the market—global consumption of OSS; otherwise, we may end up maintaining COBOL in 2100...

$5000 is enough to make a living in several countries.

On a global scale, likely less than 10% of the world's population has ever been able to save $5,000 at any point in their life, with the vast majority concentrated in high-income countries. In low- and middle-income countries, this is a rare achievement limited to a small, affluent minority.

I believe the poster was referring to a one-time $5000 grant.

It can be a recurring grant if a target OSS project continues to be highly valuable, but risky. When it loses value or is derisked (e.g. by extra funding), then grant priority will naturally move to another project.

A stable/predictable base of maybe 50k USD/year is probably more in the range where it could influence someone to be FOSS maintainer full-time.

How is this different than something like https://opencollective.com (which, for example, Actual Budget uses: https://opencollective.com/actual )

Open Collective (OC) is great! It's primarily a payments platform.

Open Source Collective (OSC, which is related to OC in convoluted ways I don't fully understand) is a fiscal sponsor of OSS projects, and is also great. :^)

Open Source Endowment (OSE), on the other hand, is a pile of money that earns interest that then gets distributed to OSS projects. So conceptually some projects either fiscally hosted by OSC or using OC as their payments platform could receive funds from OSE.

Does that help?

Edit to disclaim: I'm on the OSE board.

Good on ya.

I work on a nonprofit platform that isn't "critical infrastructure," compared to a lot of stuff, so I'd likely not seek funding, in order to avoid stealing oxygen from the lone maintainer in Nebraska.

Open Collective (really Open Finance Consortium Inc.) is a US 501(c)(6) nonprofit that runs a payments and accounting platform, providing fund acceptance and budget services to a ton of different community collectives and funding groups, making it easier to connect funders with groups that are often not incorporated.

Open Source Collective is a separate 501(c)(6) organization that actively supports funders wanting to support FOSS projects or communities specifically. They share some board members, and they simply use Open Collective to do all the finance work, while also offering some level of advice and other IP holding services: https://docs.oscollective.org/welcome-and-introduction-to-os...

Open Source Endowment is different, in that it's soliciting 501(c)(3) donations, which the OSE board and membership will use for the endowment to choose FOSS projects/communities to provide grants for.

This topic should be a FAQ page on the OSE site, especially for funders who just want to donate "to some good FOSS" without knowing where to find it. When you donate to OSC, you pick specific collectives to give to (and it's not tax deductible). When you donate to OSE, you're giving to the endowment, that the OSE Members setup policies for how/where/when to provide grants to projects/communities (and it could be tax deductible).

> Government support won't work for OSS at scale — it's too globally decentralized... We recently received US 501(c)(3) tax-exempt charity status.

If this is successful in the first iteration, I'd love to see a UK and EU based charities too. That would allow european donors to support on a gross pay basis, and may simplify grants to european nationals too. (I'm sure similar things apply in other jurisdictions too.)

Thanks, Alex. Ticketed here:

https://github.com/osendowment/foundation/issues/26

Will take some time ofc but good to plant the seed now. :)

OSE actually has it on its roadmap: https://github.com/osendowment/foundation/blob/main/roadmap....

Most likely we won't create our own subsidiaries, but will partner with local nonprofits (suggestions are welcome), which could make donations tax-deductible for UK/EU residents.

As for grants, we are totally fine with supporting European open source maintainers now. OSE has a global scope, limited only by the available payment infra and US regulations.

> will partner with local nonprofits

mmh. be very careful when choosing those. Esp. in former socialistic countries, and esp. in some of them (hint), where $$$ scheming has become bread-and-butter of the.. kind-of-former-but-new aparatchiks.. it's like an official mafia. Electrically speaking, they manage to find ways to ground and leech on any potential.. $100 or $100M alike.

otherwise - great initiative. The Commons (as of ivan ilich) need support and care in order to be .. there when needed.

Sure - we donate our own money to this endowment fund and carefully choose partners who have a solid reputation and can deliver maximum efficiency. I don't think we will have any of them in former socialist countries,except for those based in Berlin :)

There is already NLNet and other charities in the EU for funding FOSS.

Just for the love of all things, do not let this become like Wikipedia or Mozilla. The moment you start paying for irrelevant things, you lose donors current and in the future. Nothing more frustrating than those two orgs in terms of where they spend their donor funds.

I am one of https://wikimediaendowment.org/benefactors (donated years ago), and now I am totally unhappy with what is happening with Wikipedia. It has been an important lesson learned.

To keep a nonprofit efficient and impactful, it is crucial for its governance to have skin in the game; otherwise, there will be no long-term alignment of interests. More details on why and how we implement this at the Open Source Endowment: https://kvinogradov.com/osendowment

What happened with Wikipedia? I thought they run a tight optimum ship?

As long time volunteer contributor in Mozilla (I am in the firefox credits and in the monument) it is very sad to say that I agree with you.

I have my opinion on why this is happening and my explanation is that in Mozilla they lost their memory because they act as a company and not as a foundation (they have both the identities). With the layoffs they removed a lot of people working there for years that knows a lot, they lost the historic memory and I remember a lot of discussion with employees that had no idea that there was a community or that they have no idea what is a GPL license. I mean in 2018 I remember reviewing this as part of the Mozilla Reps council, https://github.com/OpenTechStrategies/open-source-archetypes that was created to help the employee to understand OSS but I think that after the managers that created moved away it was just left as it is.

Sorry for the flame but I think that when a foundation act like a corporation there are issues.

The FAQ, under "How can OSE evolve in the long term, especially in an AI-powered world?" appears to state a very pro-AI view.

I think this is hopelessly naive. The LLMs crapping out code are shamelessly ripping off open source code, sans copyright notice. It makes no sense for a foundation supporting open source to also support this massive copyright massacre.

Also, I think you're going to get flooded with requests to give money to vibe-coded crap, because if you have no skills or shame but want to make a little money off your AI-generated crap, why not try and extract money from this initiative? The curl guy showed this is very real.

> flooded with requests to give money to vibe-coded crap

And our plan is to willy-nilly give money to everyone who asks for it with no oversight or attention to other factors or human involvement. Game over. You win.

Potential issues from new tech aside, an open-source endowment is a pro-social idea, that absolutely deserves its day.

Now, setting aside ethical issues for a moment, open-sourced knowledge, writing, history, data, Q&A, and tech is essentially a prerequisite for a data-driven technology like LLMs, and if those turn out to be a net win for humanity, then we can directly trace the routes to initiatives like this one that can curate humanity's best contributions.

The curl guy is one of OSE founding donors, together with the terraform guy who recently released an open source trust management system to help with AI-generated crap: https://github.com/mitchellh/vouch

I think that AI eventually will solve technical maintenance problems, but not human-related ones: limited attention, trust, motivation issues. And we are going to support mostly "old" projects everybody relies on, not some new AI-gen stuff.

NPOs are constantly “running a tight ship.” I suspect a lot of HNers would be aghast at the limited resources available. I’ve been doing NPO work for decades.

LLMs represent a very important opportunity to force-multiply limited resources.

> It makes no sense for a foundation supporting open source to also support this massive copyright massacre.

Copyright is a fundamentally unethical concept and must die. Open source foundations should rightfully support the death of copyright, whether by LLMs or by other means.

So.... you just created a 501(c)(3) tax-exempt organization to offer grants for product development.

Yeah, this will end well.

Here are a couple salient portions of our IRS application to put your mind at ease. :^)

> In limited circumstances, the Foundation may make grants to organizations that are not described in IRC Section 501(c)(3), or to individual OSS developers, maintainers, researchers, and educators. These grants will support persons and organizations engaged in developing, maintaining, securing, documenting, or conducting research on free and open source software critical to public digital infrastructure.

> Any such grants will be made exclusively for charitable or educational purposes, with the Foundation retaining complete discretion and control over the use of funds consistent with Revenue Ruling 68-489.

[...]

> In addition to project-based grants, the Foundation will make recognition awards to individuals who have made extraordinary contributions to OSS serving as critical public digital infrastructure. These awards are analogous in structure and purpose to MacArthur Fellowships, the National Medal of Science, Pulitzer Prizes, and similar recognition programs administered by 501(c)(3) organizations.

clear and plain language on the website will do wonders compared to legal like comments with emoticons on HN

Boldly asserting that all grants will be made exclusively for charitable or educational purposes does nothing to change the character of the grant. If you're giving money to someone for commercial product development then you're giving money to someone for commercial product development ... and if that constitutes the majority of what you do then you've got a major problem.

OSE won't give money for commercial product development - it is dedicated to supporting existing highly-used _nonprofit_ and independent OSS. Some specific examples are at https://endowment.dev/faq/#grants

As soon as you start paying individual maintainers, it stops being nonprofit OSS they work on. If you direct your funds to other charities, you're only shifting the tax issue to them. If you want to give money to maintainers with no strings attached, it's basically impossible to avoid double taxation.

We explicitly explained to the IRS that our endowment plans to make awards and grants to individual OSS developers and maintainers in the US and other eligible countries. Given our limited target scope — not just any software, but critical nonprofit independent OSS — it was acceptable, and the IRS approved our 501(c)(3) status. And we plan to operate within what is described in our application.

Tax dollars to find software

lets try something new

Putting capitalists in the middle doesn't seem new, more like another place they can extract a slice of the pie.

Let's back up: The way an endowment works is that donors donate money, which goes into a more-or-less permanent investment fund. The interest from the investment fund is then used to a) fund mission-aligned programs (in our case, OSS), b) stay ahead of inflation, and c) pay operating costs.

Where are you seeing capitalists "extract a slice of the pie" here?

The README on github

"pay operating costs" is one place non-profits often find fraud. Getting the money into the market between donors and builders, now you have to pay professional investors. You don't get to 7-8% returns without equities, what happens if the market tanks?

Why not build something super minimal that requires less management and operating costs? That doesn't have the market risk at the center of it all? That doesn't have more points for fraud and abuse?

> "pay operating costs" is one place non-profits often find fraud.

If you find it here please let us know.

Can you explain the 2-3% gap between expected returns and outlays? Seems like a lot more than what is needed for accounting (based on the other main person here posting)

The explanation is simple — nobody can predict exact annual returns, and they tend to fluctuate. We aim to spend at least 5% per year on OSS grants and need to decide if we can spend more on them or should reinvest based on specific annual results. And target earnings should overcome inflation.

Reasonable answer, but this part:

> Why not build something super minimal that requires less management and operating costs? That doesn't have the market risk at the center of it all? That doesn't have more points for fraud and abuse?

could still be usefully addressed.

“super minimal that requires less management and operating costs” - that’s exactly our current setup, and always will be the target!

Now OSE has no paid employees - the team is 100% volunteers. Its Board Directors and the Executive Director are required to personally donate $1000+/year. Operating costs are close to zero.

As organization evolves there might be higher operating costs, but our commitment to keep them as low as possible.

The best long-term protection from fraud and abuse are aligned incentives through skin in the game. That’s why we legally require all people in governance to be Members ($1000+/year donation). This is an important topic, and here you can find more context on this: https://kvinogradov.com/osendowment/

I think this is really missing the point of the question. I know that it is common for endowments to be invested "in the market" - people believe that's the most responsible thing to do. But the question was about why do things the normal way? Why link up market performance of a set of investments with funding mechanisms for OSS? If you're going to be bold and try to fund something that is, in market and economic terms, quite off-norm, why do that using entirely normal systems that are at the core of a capitalist economy?

There are areas where we experiment and take risks: raising the first-ever endowment for open source, making it very lean and digital-first, relying on bottom-up funding and governance instead of large corporate donors, etc.

But all other areas should be as low-risk as possible — like accounting, legal, and investment management of a community endowment fund. We are exploring a few ideas on how to grow the fund faster than the market without increasing its risk profile, but they are complementary to a very conservative core strategy.

Besides OSE, I am a full-time VC — that's the area where investors are bold and invest in off-norm opportunities, but it lies on the totally opposite side of the investment risk spectrum. And directly mixing them does not seem like a good idea.

700k from 60+ "founding donors" = an average of 11k per person.

These are people who have net worth values in the 10's to 100's of millions of dollars and they could only give ~10k to open source software?

My goodness we're cooked. The oligarchs are so unbelievably cheap.

OSE has a diverse community of donors, and most of them are not even close to that net worth level. Around 75% of our donors contributed in $1-200k range. We think that every donation matters, and hugely respect everybody supporting open source.

May I ask what it’s the right share of net worth one should donate to OSS? And what’s this share in your personal case?

Not everyone operate on same consideration regarding money. The mere fact to be able to donate something that is a significant portion of a median income salary is already the privilege of the most wealthy, so more than that this is the very limited realms of the winner take it all game of the casino or the casino owner. For this class of people, money has nothing to do with what it represents to most people. So there is no way people from these distinct classes can understand each other in term "share of net worth", because they are the same words that refer to completely different realities.

So I know you're trying to shame my by pointing out my hypocrisy here, but I assure you that I wouldn't have made the original comment there if I didn't have a leg to stand on.

I am not a high net worth individual. With a 20 year career as an IC in the tech industry though, I'm quite comfortably middle class. But even I have donated $10k to charity in a single year many times. Many tech companies offer charity gift-matching programs. I take full advantage of those by having a donor advised fund so I can capture the full gift match amount every year and then disburse those funds as desired. When I was briefly at Microsoft, they had an oddly generous gift match of up to $15k per year. I literally put up $15k of my own money each year to capture that $30k/year into a charity account for several years.

That's why I feel like I can comment critically on this. I'm a regular dude who grew up dirt-poor and ended up modestly successful in life and even I can manage to donate amounts of 10's of thousands annually.

Compared to the much more successful executive, CEO, founder class who you mentioned were among your donors. To the kind who own houses in the Los Altos hills worth 10 million plus, those numbers seem kind of paltry.

I'm not trying to take away from the success of your organization and its mission. I congratulate you on your fundraising efforts and I know you can't afford to be critical of your donors lest those donors sour on you. But truly, it does strike me how some of the wealthiest among us, who have benefited the most from these common goods, can give so little.

Thanks for sharing the context, it's really helpful. For full transparency, I am not a high-net-worth individual either, and I also grew up dirt-poor and now am a modestly successful 34-year-old tech dude.

We are not building yet another HNWI-focused nonprofit, but a grassroots community endowment. It targets a donor like you – a regular tech industry worker who can support things they truly care about with a $1-100k donation. I think many developers love open source, don't they?

We intentionally targeted launching the endowment with a $0.5-1m initial size and many donors because:

1) It is not attractive for regular donors to support a project that is heavily funded by HNWIs. Let's say if I were able to donate to the Gates Foundation, it would be like peeing in the ocean and wouldn't matter. However, I am eager to support projects with peers among donors because it makes a difference. And that's the way to make OSE scalable and outlive all its founders.

2) It is important to maintain high decentralization of funding, which enables good community governance and accountability. Having a few outsized donors at the start kills it. Our donor Herfindahl–Hirschman index is ~1800 now.

3) As a VC (maybe a bit old school), I think that it is responsible to limit funding size at very early stages and grow it later together with achieved milestones and decreasing risks. It helps to build more efficient organizations.

Just take a look at https://endowment.dev/community - we obviously targeted notable founders at the start, but now it is a healthy mix with many everyday developers. You're welcome to join them.

Of course, OSE will look for large donations. But the goal is to raise them in balance with growing a community of smaller donors, maintaining decentralization and scalability.

They didn't get there by being nice so by definition they're cutthroat tight wads

> founders of HashiCorp, Elastic, ClickHouse, Supabase, Vue.js, Pydantic, Nginx, Gatsby, n8n, and curl

By the sound of it, we can probably expect most of the stakeholders to be less interested in critical infrastructure or anything that solves real problems for actual human beings and more interested in the kind of frivolous devops make-work that creates more problems than it solves.

Kinda up to you. Recruit your friends to join if you want a say. :^)

> Individuals contributing at least $1,000/year to the endowment fund qualify as OSE Members. Members advise the OSE board on strategic matters, such as the grant-making model, and appoint community-nominated board directors. These rights are legally defined in our membership policy.

https://github.com/osendowment/foundation?tab=readme-ov-file...

$1000 to join the discussion table, wow, that's an extremely high bar. What is the motivation of such an excluding entry barrier?

We don't require a membership to participate in GitHub discussions, but it is required to be invited to closed events and to be part of the OSE governance.

The global average salary of a software engineer is $5,906/month, and the OSE membership is $83/month ($1,000/year), so it would require donating ~1.4% of gross income. This aligns fully with the 1-2% of income that an average U.S. individual donates every year, and we are targeting more senior engineers, not the average ones.

So it is affordable for our target audience, but requires some personal commitment. To be efficient in the long term, an endowment must be managed by people who have skin in the game.

It would also be irresponsible to have people who can't or don't want to personally donate $1,000 to our cause, or people outside of the tech industry, managing a multi-million-dollar community fund focused on solving a pretty niche industry problem. Otherwise, the endowment ends up being as efficient as a typical government!

It is a community-driven initiative - we encourage developers to join as donors and help to shape it. Also, our model from the very start is about deep layers of infrastructure: https://endowment.dev/endowment/#model.

Finally, I would not say that, let's say, founders of Nginx and curl are not interested in critical infra or don't understand it :)

If someone has money, why would they be better off joining this as a donor instead of just giving that money directly to someone who's actually doing good and necessary work? Everything about this seems like it will be a waste of whatever resources go into it. Like that one time Mozilla decided to help, through one of its grants, to secure $400,000 in funding for... the Webpack project.

> I would not say that, let's say, founders of Nginx and curl are not interested in critical infra or don't understand it

Congratulations, you managed to pick out the two projects in the set that makes it merely "mostly" (instead of "entirely") frivolous.

The endowment model is complementary to direct donations, which many of OSE donors make. For example, I did this a lot: https://kvinogradov.com/algo-sponsors. Some people do this privately, and the largest public case among our donors seems to be https://ziglang.org/news/300k-from-mitchellh/

It requires choosing projects, and most people don't have time for this.

But the main problem is that such funding is simply not sustainable: corporate annual budgets for OSS are volatile, and individual donations are extremely volatile. It is not great for the critical infrastructure we all rely on.

The Open Source Endowment enables truly sustainable funding and also decreases the entry barrier for new OSS donors because it does not require choosing a specific project. A la buying the S&P 500 ETF instead of stock picking.

> Congratulations, you managed to pick out the two projects in the set that make it merely "mostly" (instead of "entirely") frivolous.

Out of curiosity, which founding donors specifically would qualify under your criteria for a "serious" project? If the founders of HashiCorp, Elastic, ClickHouse, Supabase, Nginx, and curl seem like a mostly frivolous band to you...