This is a good post overall about the difficulties and generally awful current state of open-source funding. This line especially sums it up well: I've personally done a decent amount of...
This is a good post overall about the difficulties and generally awful current state of open-source funding. This line especially sums it up well:
The dirty secret of open source is that much of it is powered by maintainer guilt.
I've personally done a decent amount of maintenance and similar work (such as helping people figure out issues) for open-source projects that I no longer really want to work on, but I know that if I don't keep doing it there's nobody else that will.
I'm glad that he stopped the experiment and seems to have dealt pretty well with the internet mobs upset about it. He still seems to feel that it's a perfectly reasonable idea though, which I think is a little strange. I understand where he's coming from that we need to try something, but I really don't think adding advertisements into users' consoles whenever they're running a tool is a good approach.
Overall, this is a huge cultural issue, and I don't know what it'll take to fix it. Like he talks about, companies have no problem paying tens of thousands of dollars for services that help them work with open-source software, but don't even consider putting similar amounts towards the people that actually maintain the software itself.
I'd love to see it become the norm for companies to do something like allocate 1% of their budget for developer salaries to go towards maintainers of the open-source projects they use. Most companies wouldn't even be slightly feasible without being able to build on top of open-source software like Linux, programming languages, databases, web servers, and more. It should really be considered natural to give something back to those projects for giving you the foundations you built on.
Honestly I think the answer lies in philanthropy and non-profit foundations. Companies won't do it because game theory combined with economics means the company that started doing so would be...
Honestly I think the answer lies in philanthropy and non-profit foundations. Companies won't do it because game theory combined with economics means the company that started doing so would be competed away by the company that didn't. You could in theory have something on the scale of a nuclear arms treaty between a bunch of companies to do it simultaneously, but that is highly unlikely.
I'm sure they can (and would) justify it like that, but it's a huge over-simplification that doesn't have much truth to it. As an easy concrete example that I know a number for offhand, reddit has...
Companies won't do it because game theory combined with economics means the company that started doing so would be competed away by the company that didn't.
I'm sure they can (and would) justify it like that, but it's a huge over-simplification that doesn't have much truth to it.
As an easy concrete example that I know a number for offhand, reddit has around 200 developers now (say it's exactly 200, for the sake of a round number). If they had 198 developers instead of 200, there's not going to be any meaningful difference to whether someone outcompetes them or not. They probably did have 198 developers until a week or two ago, and fewer before that. The difference between a 99% developer budget and a 100% one just won't really have any noticeable effect in practice. There are a ton of other factors that are more important than a 1% difference in budget.
I keep hearing people say that adding advertisements isn’t a good approach to FOSS-funding, and most people’s reasoning is, “because it’s annoying.” Why do you think ads are the wrong approach?...
I keep hearing people say that adding advertisements isn’t a good approach to FOSS-funding, and most people’s reasoning is, “because it’s annoying.”
Why do you think ads are the wrong approach? Honestly I don’t see the big deal, in fact it seems like a natural progression to me.
I'm anti-advertising in general. What advertisers want is usually the exact opposite of what users want, but when the advertisers are the source of the money they usually end up being treated as...
I'm anti-advertising in general. What advertisers want is usually the exact opposite of what users want, but when the advertisers are the source of the money they usually end up being treated as more important. Choosing advertising as your way of making money is a decision to make your users' experiences worse. The "if you're not paying for it, you're the product" saying is pithy and not totally correct, but there's a lot of truth to it too.
This is a conflict that's especially problematic when the product is a tool and not just entertainment. What if one of the advertisers doesn't like that users run other things as soon as npm finishes, so their ad gets pushed off the screen immediately? Maybe they'll request that the ad should hang the console for at least 10 seconds to force the user to look at it for a little while before they can do anything else. If that advertiser is a major one and the tool developer is dependent on them continuing to advertise, they may not be able to reject that demand.
It's just not a good path to start down, and only results in our tools getting worse.
I see your reasoning, and I appreciate you adding some more meat to it. I can certainly see how advertising might reduce a product's quality over time. That being said, I still think advertising...
I see your reasoning, and I appreciate you adding some more meat to it. I can certainly see how advertising might reduce a product's quality over time. That being said, I still think advertising is a natural progression for a maintainer of a widely-used open source project, especially as the barrier to use have come down.
From their perspective, they are creating something of value and they feel they are owed something. Advertising is one of the least-effort approaches, so it makes sense that it would be used.
I was actually surprised at the backlash. I'm very accustomed to seeing ads when I'm using "free" tools, and I certainly think creators are justified in trying to find ways to monetize their creations. If people are annoyed with his adverts, they can stop using his tool (the fact that his tool is embedded in other popular libraries doesn't really matter to me, it's still possible to stop using his tool and keep using the dependent library). Honestly I felt a lot of the backlash from developers to be a bit entitled - they wanted to use the free tool AND they wanted to keep their build logs clean.
I'm not a fan of this particular creator or of this particular tool, but it's still a tool people were using for free. You get what you pay for.
While I can sort of sympathize with the problem of funding FOSS projects, portraying a glorified ESlint config as a “Standard” style guide and linter, and attempting to monetize it, is just wrong.
While I can sort of sympathize with the problem of funding FOSS projects, portraying a glorified ESlint config as a “Standard” style guide and linter, and attempting to monetize it, is just wrong.
NPM is changing their policy to prevent ads in the future: https://www.zdnet.com/article/npm-bans-terminal-ads/ Seems like the best possible outcome of this experiment. Yes, open source needs a...
NPM is changing their policy to prevent ads in the future:
Why do you think ads are the wrong approach? Do you have an idea for a better approach? For me, I think it’s all about figuring out a way to make enterprise users pay. They’re the ones with the...
Why do you think ads are the wrong approach? Do you have an idea for a better approach?
For me, I think it’s all about figuring out a way to make enterprise users pay. They’re the ones with the willingness and ability to pay, but they won’t unless they are forced. Maybe maintainers should start forcing them?
This is a good post overall about the difficulties and generally awful current state of open-source funding. This line especially sums it up well:
I've personally done a decent amount of maintenance and similar work (such as helping people figure out issues) for open-source projects that I no longer really want to work on, but I know that if I don't keep doing it there's nobody else that will.
I'm glad that he stopped the experiment and seems to have dealt pretty well with the internet mobs upset about it. He still seems to feel that it's a perfectly reasonable idea though, which I think is a little strange. I understand where he's coming from that we need to try something, but I really don't think adding advertisements into users' consoles whenever they're running a tool is a good approach.
Overall, this is a huge cultural issue, and I don't know what it'll take to fix it. Like he talks about, companies have no problem paying tens of thousands of dollars for services that help them work with open-source software, but don't even consider putting similar amounts towards the people that actually maintain the software itself.
I'd love to see it become the norm for companies to do something like allocate 1% of their budget for developer salaries to go towards maintainers of the open-source projects they use. Most companies wouldn't even be slightly feasible without being able to build on top of open-source software like Linux, programming languages, databases, web servers, and more. It should really be considered natural to give something back to those projects for giving you the foundations you built on.
Honestly I think the answer lies in philanthropy and non-profit foundations. Companies won't do it because game theory combined with economics means the company that started doing so would be competed away by the company that didn't. You could in theory have something on the scale of a nuclear arms treaty between a bunch of companies to do it simultaneously, but that is highly unlikely.
But if Bill Gates were to leave a few billion...
I'm sure they can (and would) justify it like that, but it's a huge over-simplification that doesn't have much truth to it.
As an easy concrete example that I know a number for offhand, reddit has around 200 developers now (say it's exactly 200, for the sake of a round number). If they had 198 developers instead of 200, there's not going to be any meaningful difference to whether someone outcompetes them or not. They probably did have 198 developers until a week or two ago, and fewer before that. The difference between a 99% developer budget and a 100% one just won't really have any noticeable effect in practice. There are a ton of other factors that are more important than a 1% difference in budget.
Most of the tech companies with their massive piles of VC are already spending huge amounts of money on far more frivolous stuff anyway. Just recently, Uber said they were spending over $200,000/year on anniversary balloons in their SF office alone. They could easily find money to help support the maintainers of the open-source foundations they built their companies on.
I keep hearing people say that adding advertisements isn’t a good approach to FOSS-funding, and most people’s reasoning is, “because it’s annoying.”
Why do you think ads are the wrong approach? Honestly I don’t see the big deal, in fact it seems like a natural progression to me.
I'm anti-advertising in general. What advertisers want is usually the exact opposite of what users want, but when the advertisers are the source of the money they usually end up being treated as more important. Choosing advertising as your way of making money is a decision to make your users' experiences worse. The "if you're not paying for it, you're the product" saying is pithy and not totally correct, but there's a lot of truth to it too.
This is a conflict that's especially problematic when the product is a tool and not just entertainment. What if one of the advertisers doesn't like that users run other things as soon as
npmfinishes, so their ad gets pushed off the screen immediately? Maybe they'll request that the ad should hang the console for at least 10 seconds to force the user to look at it for a little while before they can do anything else. If that advertiser is a major one and the tool developer is dependent on them continuing to advertise, they may not be able to reject that demand.It's just not a good path to start down, and only results in our tools getting worse.
I see your reasoning, and I appreciate you adding some more meat to it. I can certainly see how advertising might reduce a product's quality over time. That being said, I still think advertising is a natural progression for a maintainer of a widely-used open source project, especially as the barrier to use have come down.
From their perspective, they are creating something of value and they feel they are owed something. Advertising is one of the least-effort approaches, so it makes sense that it would be used.
I was actually surprised at the backlash. I'm very accustomed to seeing ads when I'm using "free" tools, and I certainly think creators are justified in trying to find ways to monetize their creations. If people are annoyed with his adverts, they can stop using his tool (the fact that his tool is embedded in other popular libraries doesn't really matter to me, it's still possible to stop using his tool and keep using the dependent library). Honestly I felt a lot of the backlash from developers to be a bit entitled - they wanted to use the free tool AND they wanted to keep their build logs clean.
I'm not a fan of this particular creator or of this particular tool, but it's still a tool people were using for free. You get what you pay for.
While I can sort of sympathize with the problem of funding FOSS projects, portraying a glorified ESlint config as a “Standard” style guide and linter, and attempting to monetize it, is just wrong.
NPM is changing their policy to prevent ads in the future:
https://www.zdnet.com/article/npm-bans-terminal-ads/
Seems like the best possible outcome of this experiment. Yes, open source needs a funding model. But this was the completely wrong way to go about it.
Why do you think ads are the wrong approach? Do you have an idea for a better approach?
For me, I think it’s all about figuring out a way to make enterprise users pay. They’re the ones with the willingness and ability to pay, but they won’t unless they are forced. Maybe maintainers should start forcing them?