A new funding model for open source software • Hacker News Board

Re: A new funding model for open source software

#171

Post by caniszczyk » Thu, Jul 30, 2020, 3:19 PM UTC

jmiskovic wrote: post #96

peteforde wrote: post #63

I think that this essay is fundamentally brilliant, but doesn't come out and yell the self-evident truth: the ONLY way this ever takes off is if Github does it. And man, do I ever hope that they do. There's no other entity in the ecosystem that even approaches the role Github plays, with all due respect to Gitlab and the rest of the also-rans. It'd be wonderful if they did the same thing, but in the end, they are goi…

So we delegate complete control over open source financing channel to Microsoft, because it's most conveniant option? Sounds a bit insane. I do use both GitHub and its Sponsors feature, but I'm still hoping for open source community to come to consensus to use a service from non-profit organization like Mozilla or EFF.

There's a funding service by the Linux Foundation which projects can use as an option: https://funding.communitybridge.org

You can see other options out there too via https://www.oss.fund

In general, donations/sponsors I don't think will scale for many reasons: https://www.aniszczyk.org/2019/03/25/troubles-with-the-open-...

Re: A new funding model for open source software

#172

Post by jacques_chester » Thu, Jul 30, 2020, 3:24 PM UTC

smichel17 wrote: post #148

jacques_chester wrote: post #58

Earlier quoted context omitted.

> The reason these npm-type packages can't charge is that they're extremely fungible. I'd say it's less that they're highly substitutable (which I was I think you mean, instead of fungible) and more that they are non-excludable goods. A seller has no way to exclude a buyer from consuming opensource; that is in fact part of the definition . But without excludability, there's no incentive for a buyer to buy, other than…

> non-excludable I think the term you're looking for is "public goods", which combines non-rivalrous and the author chooses not to exclude people through legal means . Running out of time to comment before I need to get to work so I'll just link our economics page here: https://wiki.snowdrift.coop/about/economics

Yes, but with a twist. OSS in itself behaves like a public good, but most business models for OSS and closed-source software revolve around bundling to create excludability (since software is de facto non-rivalrous). That makes them club goods.

The most effective excludability is SaaS offerings, because total exclusion can be enforced easily. That's why hyperscalers make far more from OSS than companies devoted to it.

(btw: the snowdrift docs are really impressive, kudos)

Re: A new funding model for open source software

#173

Post by black_puppydog » Thu, Jul 30, 2020, 3:33 PM UTC

rowland66 wrote: post #145

black_puppydog wrote: post #104

Earlier quoted context omitted.

This might actually be the right question to ask, thanks for poking. :) From a customer's perspective, fraudulent transactions are handled in the best way possible. Here's Stripes docs on it [1]: > Customers can dispute a payment through their bank on a “no questions asked” basis up to eight weeks after their account is debited. Any disputes within this period are automatically honored. > After eight weeks and up to…

What you are describing here are SEPA direct debits. Many countries including the USA have a similar instrument. While it is free for the consumer from whom money is being collected, it is generally not free for the business that is collecting the money. So in that way, it is the same as credit cards, but the fee is a lot less.

True, but if the fees are a lot less then the question remains if they're just not worth the hassle, or if there are other reasons why adoption in companies (that I have seen) has been slow.

Regardless, an organization like github should be able to handle this, if they wanted to.

Re: A new funding model for open source software

#174

Post by ivalm » Thu, Jul 30, 2020, 4:54 PM UTC

imtringued wrote: post #89

I expected Youtube Red to work like this. Each person has its own pool. Channels should have been weighed by watch time with a minimum weight so that small channels get something as well. If I watch one channel then that channel should get the whole $10 (minus youtube cut). Instead we got a stupid global pool based on global popularity that primarily benefits PewDiePie even if I don't watch him. The compensation was…

What is the difference between popularity weighted global vs local pool.

As I understand it with global popularity.

If 9 people watch A and 1 person watches B and C equally and each pay $10 then:

Total money is $100. A gets $90, B and C get $5 each (as they get 1/20 total viewership).

This seems to be same as local pooling? Am I misunderstanding the terms?

Re: A new funding model for open source software

#175

Post by Ut_Pwnsim » Thu, Jul 30, 2020, 5:08 PM UTC

ivalm wrote: post #174

imtringued wrote: post #89

I expected Youtube Red to work like this. Each person has its own pool. Channels should have been weighed by watch time with a minimum weight so that small channels get something as well. If I watch one channel then that channel should get the whole $10 (minus youtube cut). Instead we got a stupid global pool based on global popularity that primarily benefits PewDiePie even if I don't watch him. The compensation was…

What is the difference between popularity weighted global vs local pool. As I understand it with global popularity. If 9 people watch A and 1 person watches B and C equally and each pay $10 then: Total money is $100. A gets $90, B and C get $5 each (as they get 1/20 total viewership). This seems to be same as local pooling? Am I misunderstanding the terms?

The difference is that if 9 people watch 99 hours of A's let's plays or last 2 music videos on repeat, and one person watched 1 hour of B's 20 well-researched and edited 3-minute educational videos, A-watchers put in $90, B-watchers put in $10, and under global popularity A gets $99 and B gets $1. Local pooling would mean that A gets $90 and B gets $10. This is my main complaint against Youtube Red too.

Re: A new funding model for open source software

#176

Post by kseifried » Thu, Jul 30, 2020, 5:30 PM UTC

The entities that most benefit monetarily from OpenSource are companies that sell products and services based on them. Why would they contribute to this? It's an added expense they don't have to pay. In fact a major tenant of OpenSource is that you can choose to make money or give. it away for free, but you generally have to release the source code, which is what the companies want, so again, why would they pay for it? To make a more sustainable ecosystem? Hahaha.

As for individuals funding this, Open Source developers are not cheap, to make it worthwhile (e.g. comparable to a part time contracting gig) we're talking a minimum of tens of thousands of dollars per year per developer. We're not going to get anywhere near these numbers with individuals contributing.

I write this as the guy that tried to help the OpenSource world by assigning CVEs for security issues (several thousand...), I'd have had to charge 1-200$ per CVE to make a living at this while I was doing it. That's not going to be sustained by personal donations. And even though a CVE will easily save companies a few tens to hundreds of dollars (time spent tracking all these issues when they don't have CVEs...) there's no way I'm going to get companies to pay me.

The good news is that this doesn't really matter. We've had many decades of OpenSource, there's enough good people working on this because it's their passion/hobby/day job that it's mostly sustainable on average, but specific bits may be sickly, and that's ok.

Some related listening and reading:

Episode 205 – The State of Open Source Security with Alyssa Miller from Snyk https://opensourcesecurity.io/2020/07/12/episode-205-the-sta...

Episode 185 – Is it even possible to fix open source security? https://opensourcesecurity.io/2020/03/02/episode-185-is-it-e...

Episode 182 – Does open source owe us anything? https://opensourcesecurity.io/2020/02/10/episode-182-does-op...

Re: A new funding model for open source software

#177

Post by Ut_Pwnsim » Thu, Jul 30, 2020, 5:30 PM UTC

biztos wrote: post #164

imtringued wrote: post #89

I expected Youtube Red to work like this. Each person has its own pool. Channels should have been weighed by watch time with a minimum weight so that small channels get something as well. If I watch one channel then that channel should get the whole $10 (minus youtube cut). Instead we got a stupid global pool based on global popularity that primarily benefits PewDiePie even if I don't watch him. The compensation was…

Doesn’t Patreon already solve this problem? I might have selection bias because I support an OSS dev there but isn’t that a pretty efficient way of directly supporting specific projects?

Almost, except for the low end. It would be perfect if it grouped donations and charges, and let you allocate any arbitrarily small amount to any creator. As it is, it charges fees per creator you patronize, and limits donation to being at least $1 per creator. I'd put $30/month there easily if I could spread it over 100-200 creators with fees of $0.30 + 2.9% on the $30, but the current $0.10 + 5% on each $1 donation means a lot less of my money is going to the creator (who also gets charged those fees + 5-12% for Patreon services when cashing out) than I think is useful, and that's even after downsizing the pool of creators I'd like to donate to in order to meet the minimum $1/each.

To be fair, in looking this up, I did find it's a lot better than when I cancelled my Patreon account, since at the time they were proposing $0.30 + something% even for $1 donations. But the bundling of donations was my whole point in using them, and when they clearly said they weren't going that way, that was it for me. That's why I like the Brave/flattr/"sponsor pool" approach so much more. It lets me support creators (especially web comics, bloggers, and video creators) in a much more similar scale to the advertising model, which many were on before, and for which I think we need a popular replacement.

Re: A new funding model for open source software

#178

Post by biztos » Thu, Jul 30, 2020, 6:28 PM UTC

Ut_Pwnsim wrote: post #177

biztos wrote: post #164

Earlier quoted context omitted.

Doesn’t Patreon already solve this problem? I might have selection bias because I support an OSS dev there but isn’t that a pretty efficient way of directly supporting specific projects?

Almost, except for the low end. It would be perfect if it grouped donations and charges, and let you allocate any arbitrarily small amount to any creator. As it is, it charges fees per creator you patronize, and limits donation to being at least $1 per creator. I'd put $30/month there easily if I could spread it over 100-200 creators with fees of $0.30 + 2.9% on the $30, but the current $0.10 + 5% on each $1 donation…

Thanks for looking it up. This was news to me, I always assumed Patreon charged whatever fee on the gross amount I pay every month.

Re: A new funding model for open source software

#179

Post by Ajedi32 » Thu, Jul 30, 2020, 6:45 PM UTC

biztos wrote: post #164

imtringued wrote: post #89

I expected Youtube Red to work like this. Each person has its own pool. Channels should have been weighed by watch time with a minimum weight so that small channels get something as well. If I watch one channel then that channel should get the whole $10 (minus youtube cut). Instead we got a stupid global pool based on global popularity that primarily benefits PewDiePie even if I don't watch him. The compensation was…

Doesn’t Patreon already solve this problem? I might have selection bias because I support an OSS dev there but isn’t that a pretty efficient way of directly supporting specific projects?

Patreon has the same issue described in the OP: the marginal cost (both psychological and financial) of supporting additional creators is too high. Since YouTube Premium distributes payments automatically based on watch time it doesn't suffer from that issue.

Re: A new funding model for open source software

#180

Post by riidom » Thu, Jul 30, 2020, 10:16 PM UTC

dimitrios1 wrote: post #70

I like the idea but instead of pools going arbitrarily every which way or to the discretion of the donator, it should be automatically applied across that open source project's dependencies as well. If I am open source software A, and I make ample use of open source software B within my project, your pool donation should automatically apply some to B, relative to it's usage in the project.

I had a similar idea to this while reading the article. Basically, people donate primarily to end-user facing projects, like they do anyway (if at all). Each dependency gets 10% of the donation, which gets capped at 50:50. If there are more than 5 dependencies, these share the 50% among them.

So for example, in case I described it bad: Project A (1 dep.) gets 90%, the dep. 10%; Project B (5 dep.) gets 50%, each dep. gets 10%; Project C (10 dep.) gets 50%, each dep. gets 5%.

Now for the loopholes:

How to track dependencies? Well, I think the fear of public-shaming and pulling back of donations in case of bad acting would stop the end-user facing projects from lying. This donation system (ideally run by a trustful brand like Mozilla etc.) has a website, where you can look up projects and the dependencies they list. Since being open source is a requirement, claims can be checked.

Should all dependencies be weighted the same? The massive UI library, and the tiny bare-bones A*-implementation, or the NPM micropackages? This needs a metric that can't be rigged, so things like number of contributors, LoC or number of commits are not really sufficient.

What about MIT-licensed libraries? They may get used a lot by proprietary projects, but these can't receive donations in first place, so there is nothing to share to the libraries. Now the opinions on MIT are kinda mixed, but simply saying "your fault, pick a viral licence" is too easy, imo.

__

Footnote: How the [user]--donation-->[end-user facing project] chain works in first place, is not really part of my idea, I wanted to focus on how to get the money to the people whose project names nobody knows (until things like Heartbleed happen).