Hacker News Reader

This is an unfair go at the poster. He said "I'm not arguing that npm install logs should be packed full of ads (it shouldn't)."

That's clear as can be. He's asking the same question I am, how do we get sufficient funding for these projects. Re your patreon suggestion, I thought we recently had an article where someone got little to nothing on that.

> or getting hired by FAANG

So one way to support foss software is to get hired by a closed source company. Kind of defeats your point.

This is the problem right here, "I want but I'm not personally going to pay". We get ads (funding model of the modern web) or foss remains underfunded while non-foss companies get the cash they need. The problem is human nature. And yes, I've personally opened my wallet for them, and will again.

>This is an unfair go at the poster. He said "I'm not arguing that npm install logs should be packed full of ads (it shouldn't)." That's clear as can be.

Please reread my reply to OP carefully and notice I did not say he endorsed NPM ads. Instead, I specifically disagreed with his suggestion to redirect the discussion away from the NPM ads to the general topic of open-source funding.

>He's asking the same question I am, how do we get sufficient funding for these projects.

To be clear, that's a different question and I wasn't even attempting to answer it. I mentioned funding like Patreon as examples of social acceptability and not as examples of income sustainability.

>Kind of defeats your point.

My point was a narrow one : others are criticizing the unwelcome ad solicitation and that's a correct area of focus.

The other question of how to get sustainable open-source funding is also an important topic -- but that doesn't mean it forbids discussion of unacceptable spam/ads. They can be 2 parallel discussions.

you said

> However, if the concrete implementation of trying to get money is unwanted and unexpected ads, then the correct focus of discussion is the criticism of that ad delivery method

But he'd already said, prior to that

> I'm not arguing that npm install logs should be packed full of ads (it shouldn't).

He'd already said no to them; you're both in agreement. There's nothing to argue about.

Having agreed that nobody wants ads, the next step is to determine how to not have them (and their nastier big sibling, telemetry and malware, which will follow) by providing funding.

(edited for brevity)

You’re missing the point and illustrating it further. The comment was garbage because it derailed the entire discussion around terminal ads to try to direct it to FOSS funding.

It doesn’t matter if they both agree that text ads are bad. It’s basically a “think of the children” derailment except it’s “think of the poor foss dev”.

I think you are making an extremely valid point; i often see the same “think of the children” argument made when discussing advertising and the web. Likewise, no one is saying websites shouldn’t have an income.

If you're only talking about how you don't like the solution; and refuse to talk about the underlying problem; you're going to keep seeing the bad solution again and again and again.

Here everyone says they don't like ads; and then are shocked that people want to talk about why ads are showing up and what to do about that.

Ads show up regardless of funding. Cut the bullshit. Look at how many ads are on cable television and even pay for news websites.

Please stop having "you said he said" discussions. If someone clarifies their position that they meant to say something else than you read, please accept it in good faith and don't see this as an invitation to hark back to their previous messages.

This comment is not aimed at you specifically, but everyone in this comment thread who engages in this pointless arguing.

Since when are FAANG companies considered to be closed source? Have any of them snatched up maintainers to continue development behind closed doors?

There are examples of large corporations hiring maintainers to work on open source and keep those projects open source. I believe pypi is an example of that.

I am fairly certain that Apple has hired people who previously worked on permissively-licensed projects to work on the closed-source fork.

I'd like to see that backed up, really. There are certainly criticisms to be made of how Apple has handled their open source projects (slow to update and push out new releases, not great at cooperating with upstream projects when Apple has forked them, typically cryptic Apple communication practices). But if Apple has taken open source projects, created closed-source forks of them, and hired original developers of and/or major contributors to the original open source project to maintain and develop the closed source fork without contributing back to the open source original, none of them immediately come to mind.

Apple certainly hires people from their open source projects and keeps closed-source forks. Most of the changes make their way back to their original projects, though, but the caveat is that this only seems to happen if the project is used heavily outside of Apple and after the feature has "shipped" (which may be a while after the code was actually written).

I remember finding an article on HN about a developer for FreeBSD going to work on Mac OS[1], before returning to work in open source. Isn't the Mac OS user space an example of what you describe?

[1] https://www.wired.com/2013/08/jordan-hubbard/

I don't think I'd say it is, no, because Hubbard wasn't hired to work on a closed-source fork of FreeBSD at Apple but rather as an engineering manager. AFAIK, Apple did not make any closed forks of major BSD-related work.

I got direct quotes from Apple employees about how their employment has banned them from doing anything related to GPLv3 code in their free time, specifically GCC.

I don't want to mention them in a public website in case that gets them in trouble.

Well, depending on the employee, it would make sense. Apple funded a large part of clang/llvm, which many assume was specifically to avoid a GPL compiler. You'd hate to accidentally pollute a BSD(-ish) project with GPL code, if your primary purpose for funding the project is to avoid GPL code. Not that anyone would purposefully do such a thing, but it would be important for Apple to even avoid the appearance of anything like that.

That doesn't surprise me. The GPL, especially v3 and AGPL, are specific weird cases at a lot of companies -- sometimes because they want the freedom to close of forks, but often because they're convinced the GPL/AGPL could "poison" their work and force them to open previously closed source. (This turned out to be a problem for us at RethinkDB, which was AGPL-licensed.)

They've certainly hired people to work on open-source projects and keep them open-source (e.g., LLVM, CUPS). What closed-source forks have they made?

Apple keeps internal forks of projects where they can do so legally; it helps them prevent leaks and lets them add new features quicker. These changes often do end up getting merged back in, but this is usually after they have shipped.

Since always? Their core business is developing proprietary software. It's impossible to git clone and say, fix any of the popular complaints.

You might be confused because they build on top of a Free foundation, and do contribute changes back to those layers. But the bulk of their activity is plainly closed, especially the software directly associated with their brands.

> That's clear as can be. He's asking the same question I am, how do we get sufficient funding for these projects.

We don't. Only a fool would expect to make a living wage giving something away for free.

From the beginning FOSS has been either interested individuals or employees at corporations with a vested interest. There's no reason we should expect that John Q Developer independently working on a bunch of NPM modules to be compensated by the world at large.

> There's no reason we should expect that John Q Developer independently working on a bunch of NPM modules to be compensated by the world at large.

With all due respect, I feel this is a lack of imagination. We could have said the same thing about soup kitchens, or basic income, or social welfare, or anything else that doesn't cleanly draw profit.

So yes, it doesn't work. At least not until we demand it does and distort the legal, social, and cultural frameworks around it.

As one solitary example, what about a shift to using copyfarleft licensing that only allows MIT-level freedom to nonprofits and co-operatives, while for-profit orgs pay fees instead of just leeching. This has the added bonus of shifting benefit to platform cooperatives, which governments are already starting to ideologically favour as we all see the gleaming metal edge of platform capitalism.

There are a hundred more options if we believe building the commons is better than milking it dry.

> With all due respect, I feel this is a lack of imagination. We could have said the same thing about soup kitchens, or basic income, or social welfare, or anything else that doesn't cleanly draw profit.

Those are all driven by charity for the destitute. It’s not meant to be a patronage paid to starving artists. On the contrary the able bodied are expected to not draw from that well if they can provide for themselves so that more remains for the truly deserving.

> So yes, it doesn't work. At least not until we demand it does and distort the legal, social, and cultural frameworks around it. As one solitary example, what about a shift to using copyfarleft licensing that only allows MIT-level freedom to nonprofits and co-operatives, while for-profit orgs pay fees instead of just leeching. This has the added bonus of shifting benefit to platform cooperatives, which governments are already starting to ideologically favour as we all see the gleaming metal edge of platform capitalism.

There’s no such thing as partial BSD style licensing. You’re either all the way there or you’re not. Favored status to non profits is no different than discriminating from “evil” usage.

I’m not saying one can’t do that, just that there’s no half measures. I make it a point to contribute to BSD style software that I actually use and explicitly do not contribute to “shared source” bullshit.

> There are a hundred more options if we believe building the commons is better than milking it dry.

I don’t see anything going dry. The system has worked fine till now. What’s changed is the unrealistic expectation of hand outs for something you’ve already agreed to give away for free.

I would like to disagree with your point on basic income. It’s usually called _universal_ basic income. Precisely to clarify that it is not a “charity for the destitute”, it is absolutely meant to be for everyone.

And for the larger discussion on building the commons I happen to think UBI is an important component. Focusing on the amateur work being done by the civil society at least, an UBI could be precisely the thing needed to free up time for cultivating the commons.

Now, granted, most UBI proponents do argue from a solidarity angle, and do think of it as some kind of charity. But I’ll offer an alternative view point. To me UBI is the foundation from which private property can be justified. Locke argued that commons, such as land, can only be taken as property when there is no contention, which is clearly not the case for anything having a market value. The just way to solve that conflict could be for the owner to simply pay the market value in rent to the commons, as compensation for the exclusive rights. Such payments could be divided as as an UBI. And I suspect it would also result i a pretty handsome amount. Enough to fund a few open source projects at least.

> I would like to disagree with your point on basic income. It’s usually called _universal_ basic income. Precisely to clarify that it is not a “charity for the destitute”, it is absolutely meant to be for everyone.

In the credible UBI plans I've seen, the income is Universal in the sense that "everyone gets a check", but in practice that income is paid for by those with middle or higher incomes, such that for many earners that "income" is a net tax (or neutral). "Charity for the destitute" is a bit hyperbolic, but its absolutely income supplementation for the less well off at the expense of those better off.

In case it was missed. My take in UBI outlined above would avoid this. Even better, I’m hopeful we could abolish all forms of income tax.

I agree with the sentiment that shady things shouldn't be excused because "I'm an open source developer and I need money". You have a point there.

Although:

> decides to embed advertising such as "Try LINODE for 30 days!" in "README.md"

Are you trying to say that this is also not acceptable? Our company sponsors a few open source npm projects, and all of the projects we sponsor have our logo in the README.md - is that really not socially acceptable?

>we sponsor have our logo in the README.md - is that really not socially acceptable?

In the d3 example I tried to exaggerate the hypothetical intrusiveness of ads by embedding it in all the js source files.

In your specific case, if nobody is complaining about your ad inside of a single "README.md", I'm going to presume it's "acceptable".

There is a difference between a company logo and "Try Company PRODUCT for 30 days!"

I imagine the latter comes with a link, perhaps even a tracking one, to sign up for the service.

Not OP, but I still think that's acceptable. README.md is one of a very small number of places that open source developers can reach their users. Most of the time I only read it once, so if there's an ad there, or a link to their other products, as long as it's not a GIF and they didn't find some way to embed a tracker in README.md, I'm cool with it.

It's about as relevant as advertising gets, and it's not coming from a pool of clickbait ads. It's a developer using their open source work to generate interest in the work that pays for it.

As an open source developer, I would refuse to "reach" my users in such a tacky way as selling ad space in my README. You're correct that the README is one of the few touch points with users, and I wouldn't dare waste that on something I consider so disrespectful.

Oh, I'm not talking about selling ad space. I agree that what Funded did was misguided at best. But what my parent was responding to was the idea of even putting a little ad for a paid product by the same developer or company.

If JetBrains, for example, wanted to remind me on the README for Kotlin that they have an IDE that works well with the language, that doesn't offend me.

An affiliate link, I suppose. Which is annoying, as these are practically spam.

> perhaps even a tracking one

Tracking the campaign? Yes, probably. But tracking you? How do you expect that to work?

Making ads contextual to the place they are and tracking the campaigns to discover what works and what doesn't is how it should be done.

I presume the user is assuming you could put a tracking pixel in the readme and have it rendered, the same way the build badges work. Github is, of course, one step ahead and hides those behind camo.githubusercontent.com.

> Are you trying to say that this is also not acceptable?

Not OP but I think it's acceptable. I'm happy to see my favourite free (as in freedom) projects' developers being supported. If someone doesn't like it and if it's truly free software then anyone can fork it to remove that ad and we don't have a problem. That's how free software works, if you can't, then it's not free software. If one just won't fork and opposes what a free software project does then that's just entitlement.

> You're (possibly unintentionally) distorting/diverting the issue. Nobody is criticizing open source maintainers for trying to get funding in an abstract sense.

Yes they're very good at providing speech and debate lip service about wanting to sustain OSS devs, but then when we actually want to do anything except hold out a tip jar they treat OSS devs like trash. The person that wanted to throw people in prison didn't seem particularly interested in our ability to get funding https://twitter.com/kyledrake/status/1166801737534984192

if the concrete implementation of trying to get money is unwanted and unexpected

Every method of funding open source either doesn't work or is unwanted, so arguments like "get paid, just not this way" are disingenuous. When that guy tries a different approach instead of ads, people will be attacking him just as much.

It's time to move the discussion on to "all possible funding methods are unpalatable, so what's the least bad method?"

Open source project funding is a major open problem, which is something many people agree on. It can't sustain if we don't solve that.

What I don't see is the fairness argument. Developers have legal means to restrict commercial use of their software. The community has for decades had a conversation about licensing and none of this is new territory.

These developers chose to volunteer their time to corporations. If they don't like that choice they should change their behavior not make demands of others.

Otherwise it's just the high tech equivalent of the common street scam where somebody hands you an object and then demands a donation if you take it.

Oh sure, but Industry has also browbeat people into believing the GPL is completely untenable.

I've yet to see an original monetization suggestion in these threads. People have tried it all. Companies and devs prefer to wait for a poor sap to make a free-as-in-beer version and then stiff them on tips.

Not sure it’s a good one, but how about this:

A hub of software, like github or npm could acts as a general mediator of funding. A new license could be designed such that entitles above a certain size would have to donate a percentage of revenue to the hub, while also reporting which software they which to license (this only acts as a kind of voting for where to place investments, the donation size is fixed). The hub in question could then spend the income on the various projects taking popularity in account for prioritzation.

> These developers chose to volunteer their time to corporations.

Not so, these developers chose to volunteer their time to benefit the commons, and corporations occupy the same spaces once in a while. Whether or not there are abusive companies who stake out those projects for rent-seeking, to build walled gardens of their own, looting the commons making money for themselves and giving nothing back, is really not a function of the person who did something good for everyone's benefit.

We shouldn't be asking the person who is doing the good thing to change, friend! Reciprocity is an evolutionarily acquired trait, and while you can't count on favors to be repaid all of the time, anyway this is not a requirement in order for us to benefit from reciprocity.

Students of psychology have understood this as one of the favorable characteristics of human behavior which enabled us to survive up to this point. The idea that you can give someone something and receive something bigger in return, while both parties benefit, has been a part of community building for as long as humans formed communities. I would like to share this Sandi Metz talk about it in case you have an hour and want to hear more, I'm not the student of Psychology and didn't make this stuff up myself:

https://www.youtube.com/watch?v=VzWLGMtXflg

> You're Insufficiently Persuasive by Sandi Metz

> Not so, these developers chose to volunteer their time to benefit the commons, and corporations occupy the same spaces once in a while. Whether or not there are abusive companies who stake out those projects for rent-seeking, to build walled gardens of their own, looting the commons making money for themselves and giving nothing back, is really not a function of the person who did something good for everyone's benefit.

It actually is, that's my point. The developers knew there were licensing arrangements that guarded against that possibility, and they chose licenses specifically designed to allow such looting.

Reciprocity is great. But if you license your code in a way that says "Feel free to take this and make money off it and give nothing back" when there are other options, then you made that decision and there is no obvious fairness argument that somebody should be punished for doing what you granted them license to do.

The point of the linked video, at least the relevant part about the street scammers that hand you something and then ask for money, is that we actually banned that kind of behavior in public spaces because it was exploitative.

People have evolved to trust in reciprocity, because it's to our mutual benefit as a society that reciprocity remains a thing, and behaviors which are pathologically exploitative of this evolved trust in reciprocity are fully in the wrong. We decided it and made laws about it, something like 30 years ago. I learned this watching the keynote myself, and was surprised (as it actually hasn't stopped, in spite of legal protections which you might assume put a permanent end to the practice, to the contrary there are still monks handing out little plastic bracelets outside of the Smithsonian in DC, and no shortage of people who are not wise to it, with the $20 to spare.)

If the next generation of developers can't anymore trust in reciprocity and they have to decide on non-free licenses as a result of these companies which plainly don't understand reciprocity, we will all have lost something profoundly important. (And if we assume these companies and their behavior is purely exploitative now, what makes you think a legal machination like non-permissive licensing is going to have more success at getting them to stop doing that? You might have more avenues for recourse, but at what cost...)

It's not about fairness or punishment, it's about protection of our shared mutual benefit via social cues, and we can exile or something like exclude them from polite society if they are not well-behaved. From the receiving end it might look like a punishment, but I prefer to think of it more like as "corrective prodding" or "defensive posturing," and if it works the bad behavior will change, or if it doesn't, then hopefully at least the blast radius can be well-contained.

There are corporations which have learned to behave more thoughtfully and in harmony with OSS, who made a point to be aware of their community footprint, and sure plenty of such individuals too. Those who are not well-behaved can either hopefully see the light, or maybe there's no hope remaining, they will totally take over, and complete the tragedy of the commons. I submit humbly that we should not degrade the commons though as a response to their influence, because even with the bad actors around and their bad behavior, the facts show that our innate understanding of reciprocity is mostly still a beneficial trait, worthy of keeping around.

> or getting hired by FAANG

But that way, you're still likely funding your project through ad scumbaggery, you're just adding a layer of indirection.

I disagree with Patreon and FAANG as possible socially acceptable routes, and have no problem with ads in the console. (Grant's are clearly OK). So I guess that might be subjective?

I do think that instead of everyone putting in an "ad dependency" is not the best way to go about it, but to at least directly negotiate with cloud companies to do it would be awesome.