Hacker News new | past | comments | ask | show | jobs | submit login
100 million dollars to reshape the economics of the web (foundation.mozilla.org)
495 points by lftherios on Sept 16, 2019 | hide | past | favorite | 269 comments

None of these web monetization efforts address the fact that the pool of [all consumers' discretionary income they'd be willing to spend] is likely a small fraction of the size of [all commercial companies' marketing budgets]. No matter how much you reduce the friction for having people make small tips to websites, there's just significantly less money to go around, so it can never replace advertising revenue. The only way that would change is if online advertising went away overnight, and instead of companies redirecting that to other forms of advertising, the prices of all consumer goods went down so it stayed in our collective pockets.

This kind of seems like a troubling insight at first glance, but it doesn't pass basic sanity checks imo. You could make the same exact argument about any other type of industry. Watch:

>The [pool of discretionary income people have to spend on books] is a small fraction of [all commercial companies' marketing budgets], therefore it will always be more profitable to run free bookstores that try to make money by putting ads inside the dustjackets and all along the shelves.

But this obviously isn't true - Barnes and Noble is a thing. Or you could even consider Kindles and the ebook market examples of how this isn't true. It can be far _superior_ in terms of revenue to offer a product that people pay for directly, as opposed to baiting them into a space with ads in it.

I think the mistake came in when you compared all of the marketing budgets combined, (call it M), with all of the discretionary income combined, (D), and took it as axiomatic that D << M. But... how could that possibly be true? Most companies will not be operating at a loss, which means their annual marketing budget will be some fraction of their gross revenue. From what I can tell from some googling, at least ~60% of that revenue, averaging over all corporations, is coming directly from... consumers' pocketbooks.

Some more googling tells me that marketing spend from US companies was ~$200 billion in 2018. US consumers spent $14 trillion. So the whole premise strikes me as questionable. Am I missing something?

I think what you're missing is that discretionary spending is not all spending, and marketing budgets are a percentage of all spending. When I go to CNN.com right now, I see ads for Tide pods and Toyota SUVs. That ad spend came from our supermarket and car payments, which you can't redirect to tipping websites unless you want to stop eating and driving to work.

My argument is basically that no web monetization plan that involves redirecting some of our discretionary spending towards paying websites can make up for the advertising revenue they currently get, because that ad revenue is funded by not only a portion of our discretionary spending, but a portion of ALL our spending, and the non-discretionary portion is much larger. Tide and Toyota are always going to have more money to spend on advertising than Bobby and Jill have the discretion to spend on sponsoring websites.

It's all just spending though. Tide is trying to get you to pay a little extra for the name brand over the generic. Instead of a Toyota SUV you could have picked up a used beater - or kept driving the car you have for a few more thousand miles. Netflix wants you to shell out extra "discretionary" money to watch some TV shows you don't need to see.

And millions of people do it, because they want to. If it cost money to read a a website, then people might do that too - if they deemed the content worthy of their hard-earned dollars. Millions of people already pony up $1000 for the latest smartphone so they can read stuff online at all hours of the day. I really don't think the web creators should be this worried about whether or not people actually want their stuff. They desperately want it; in fact, most people are probably literally addicted to it.

Of course, maybe you're trying to draw the distinction between paying for a good and voluntarily making a donation. If that's the case, I'm with you - the latter idea is a loser. But if popular websites just made you pay to use them, and the amount was tiny, and the actual process of paying was completely frictionless and unnoticeable in terms of the browsing experience, then people would totally do it. I think a lot of people think they learned this fundamental truth that people don't want to pay for things to read on the internet, when what they really learned is that people don't want to be interrupted and asked for their credit card details. But if you can make that process seamless, people will happily fritter away thousands clicking through links.

> But if popular websites just made you pay to use them, and the amount was tiny, and the actual process of paying was completely frictionless and unnoticeable in terms of the browsing experience, then people would totally do it.

People would maybe do it, but ads wouldn't go away because there's x times more money to be made in the ad business than people micropayments.

I think on demand TV works great, you want to watch a movie hit the remote and the cost will be part of your monthly bill.

If something similar existed for the web it would be great

It's highly dependent on implementation. It's too much mental load to determine if that article you just read is worth $0.02 or not.

Brave has a model where creators are automatically tipped, based on usage, which makes more sense imo.

The main issue with Brave's funding model is that it relies on consumer goodwill. While people might be willing to spend $0.02 to read a news article, they're less likely to allocate $10/month to funding online content creators "just because". There has to be some sort of incentive for most people to spend their money; if they can get the same experience for free, only a small minority will ever bother to pay.

So the question is, how do you provide consumers with such an incentive without charging for the content itself (as that can be problematic for the reason you mentioned), without offering "an ad-free experience" for paid users (as Brave blocks ads by default regardless of whether users pay or not), and without requiring users to incur the mental load of trying to decide whether the paid experience is worth it over the free one?

A difficult problem to say the least.

I often hit the free limit on say articles on New York Times and similar sites. After seeing the article title I am often interested enough to click on it to read it. But the count of free articles is full so I can't read the article. Now if it said "If you want to read this article there will be $0.50 charge in your next month's internet bill, I would probably go ahead and click and pay it in the next bill.

Problem is I don't want to open an account with just every possible newspaper and magazine on the web. What is needed is some common mechanism for most if not all content providers.

Maybe $0.50 is too much. Maybe $0.25 would be more like ok.

Your point about having a single account is correct, but the sheer amount of decisions when browsing the web would quickly become overwhelming.

Say that in an hour you browse two articles from the NYT, one from the WSJ, five Wikipedia pages, one click on a BuzzFeed listicle, and one blog post from Joe's blog.

Is the NYT article worth $0.50? $0.25? What about the WSJ one? And surely Wikipedia is worth as much as a NYT article. But the BuzzFeed one is definitely worth less. Maybe $0.05? It was moderately entertaining after all. And what about Joe's blog?

Having to do dozens of these micro decisions per hour is exhausting, and is the whole point behind having a system that allocates funding automatically. It's not optimal, but it avoids decision fatigue.

I'n thinking that the content-provider obviously would set the price, I would decide whether to click if it is 25 or fifty cents.

I don'think it would be more exhausting than being in a bar and putting coins into the jukebox.

Automatic funding sounds interesting, was it explained in the article somewhere? Or is the point that we should use 100 million to develop a working automatic funding algorithm since no such thing exists yet?

why is that different from say...a movie youre going rent on apple TV or in the theater?

I think the big thing is trust in the market. Consumers tend to be conservative in which parts of the market they trust. They don't want to pay for something if they have no idea if they're going to get their money's worth. With existing markets, or any market that lots of people are using, people trust that it's okay so they join. With something new, are you going to pay money to access something if you don't know if it's any good? Moving towards free+ads is always easier than towards payment up front.

That said, Netflix is incredibly successful with something new, but that also took years. Netflix worked hard on the reputation that they had the best, most interesting new TV shows that you couldn't get anywhere else. But it takes years to overcome that inertia.

> Tide pods and Toyota SUVs

> eating and driving to work.

This confused me for a second before I realized it was intentional. Good one

Inserting humour in a HN comment and making it highly upvoted is an artform in itself.

General shoutout to anyone who's less lazy than me: has there been any good research on whether or not advertising increases discretionary spending, and if so by how much?

The Tide example here makes me feel a little bit weird, because Tide pods in particular are a pretty bad deal for laundry detergent. If people are buying Tide pods instead of generic laundry detergent, it seems to me that the advertising isn't just shifting money around, it's actually increasing spending. But it could just be bad instincts, I don't know of any actual real research on the topic.

Another way of phrasing this -- if Apple stopped advertising iPhones, some Apple users would shift over to Android, which (ignoring cost differences) theoretically would just mean they gave the same money to someone else. But would there be another contingent that stayed on iPhone and just stopped buying new hardware every year? How large would a contingent like that be?

A lot of advertising is not meant to increase discretionary spending, but to make you pick product A over product B when the moment comes to choose over alternatives.

At the very basic level of it, advertising is giving out the information that your product exists, because otherwise, nobody would even know its existence.

Your example for Apple is very good, it's very highly probable that some persons only switch to newer iPhones when they hear the news it exists and see the keynotes/articles about it, even though there is nothing wrong with their phone, and they may not be thinking about changing it.

> The Tide example here makes me feel a little bit weird, because Tide pods in particular are a pretty bad deal for laundry detergent.

For my wife, the pods are a lifestyle upgrade over the regular powdered pour out detergent. The convenience and effort saved justifies the more expensive packaging.

I'm a firm believer in adopting new processes or products that reduce the amount of time required to perform my daily functions. They pay off in the long run.

I think advertising increases discretionary spending, it just redirects it.

When smoking ads were banned, profits for cigarette companies went up, because they didn't have to spend money on marketing anymore. Addicts still kept buying them anyway. Previously, smoking ads were to get people to buy your brand rather than some competing brand.

Here's some of Google's positioning on it - think of it as an arms race: https://www.thinkwithgoogle.com/marketing-resources/micro-mo...

I’ve always figured they wouldn’t throw so much money at it if it wasn’t working. I assume the advertising makes them massive profits.

I’m sure there are pretty of studies out there.

But for any given business the maths work out differently. Building a dedicated audience in a niche who are willing to pay for a thing directly is easier than wrestling with the advertising industry to try and extract enough ad revenue to keep going.

> no web monetization plan that involves redirecting some of our discretionary spending towards paying websites can make up for the advertising revenue they currently get, because that ad revenue is funded by not only a portion of our discretionary spending, but a portion of ALL our spending, and the non-discretionary portion is much larger.

It kind of sounds like ads are like a pollution on the web, and we should just ban them.

What you describe is a lot like the pollution due to over packaging many products in the food industry. You can say consumers shouldn't buy it, but consumers as an aggregate can't handle that and if you want to solve the problem, it's to put less packaging on the shelves.

Similarly, if what you describe is true (and it sounds true, because how massive the internet ad industry is, indeed to me never seemed to quite add up to what people do online), then no matter what sort of monetization schemes we come up with for Internet content, most industry everywhere is still going to pollute the Internet with ads, simply because the content producers want more money, the consumer doesn't have it, but the industry marketing budgets do.

I think this is partly due to "content producers want more money", because if it's never enough you can wreck any good thing. But the other part is that "the industry" can apparently stuff the Internet full of as many ads as they want without actually being bound by how much the consumer wants to "pay" for the content via ads or (if we figure out micropayments or something) money. Because their spending budget dwarfs that of the consumer.

So, just like I always suspected, all these ads are there because the industry wants to advertise on the Internet, not because it needs to be there to pay for the content.

Content automatically becomes advertising when it exists by the virtue of a marketing budget.

Turns out this whole "your data for our content" deal, that consumers are supposedly choosing for, isn't the relevant transaction at all.

You see pretty much the same problem with outdoor advertising and store front signage. The industry has way more than enough money to spend on advertising to ruin the urban scenery with ads and giant billboards. Where I live we have regulations to keep that in check, but when I go to other countries it's really ugly. It's quite bad in the USA, but at least that country hasn't existed for very long. The saddest thing is in older countries where you see historical buildings being half-obscured or even have the billboards bolted on them. Fortunately due to tourism, there are some regulations.

Anyway, the point is that our Internet is being ruined with ads by the marketing budgets of the industry, not because the ads are necessary to support the content (they never were--there was content on the Internet before advertising, lots of it too), but simply because they have the money to put as much ads as they want onto our Internet.

It reminds me of certain cafes where they play bad commercial radio. You paid for your coffee, to sit there, but you're still subjected to over-loud ads every 15 minutes. This is not because you didn't pay enough for the coffee, it's because whoever paid for that ad has the resources to pollute the soundscape regardless of whether you pay or not.

To add to your argument: Spotify has 206 million users - almost half of which are paid subscribers.[0] This describes exactly the point of contention since free users are financing spotify through ads instead.

I think it is also plausible to assume that among the paid subscribers a higher margin are primary users of spotify for music consumption, while free users have a higher share of secondary users (relying more on vynil, CDs, limewire).

The dark side then again for example are news sites - very few of which seem to be able to finance themselves well via subscription services.


Your argument breaks down when you consider the thing that allows newspaper/web ads to be profitable is the cost of delivery effectively being reduced to zero, whereas a book still has massive printing costs that cannot be scaled as well (thousands of the same books vs. millions of the same newspapers/page views)

B-b-but what about ebooks?

Ebooks are an awful experience both buying and reading.

Alternatively if advertising became significantly less effective companies would spend less money on it. That's the upside of Ad-blocking which could theoretically make a huge impact.

> Alternatively if advertising became significantly less effective companies would spend less money on it.

What's the strategy for making advertising on Google and Facebook (+Instagram) - where the majority of all online advertising outside of China occurs - significantly less effective?

Google controls search, Android, Chrome and the Android app store (along with of course YouTube, Google Maps and Gmail, three other giant services). Facebook has its own massive walled off kingdom that exists nearly entirely in apps where they can heavily control ad blocking.

Google nearly doubled its sales in three fiscal years, from $74b to $136b. Something that big doubling in that span of time, is extraordinary. They'll probably hit near $160b in sales for 2019. I'll take the bet that a giant that large, entirely dependent on advertising, will continue to figure out how to insulate itself from ad blocking by controlling the territory around it.

Your premise will always be a non-starter so long as the ad giants control their own sprawling acreage (which they will fight to the death to retain, because it is life or death to them). The ad giants have made sure to control their own destiny, they saw the ad block threat plainly a decade ago or more and moved to stay out in front of it.

> Facebook has its own massive walled off kingdom that exists nearly entirely in apps where they can heavily control ad blocking.

True, hence the rise of DNS based content blocking like pi-hole. Sure, it is trivial for apps to workaround it by using a custom DoH impl themselves, it hasn't been done yet.

Another possibility is to patch the apps, ref YoutubeVanced

Yet another possibility is to run an app within a user-land emulator on the device and block content and deny requests. See: VirtualApp

I'm not sure where you got this idea.

Ad-blocking might hurt in some future with hypothetically high enough percentages of traffic.

Right now they make ads more effective (marginally). By self selection, a population with relatively low click-rates has removed themselves from having any money spent on them. This helps Google and Advertisers since right now, the volume of available impressions is plenty high, the name of the game is targeting better.

If anything, ad blockers just hurt select publishers who can potentially see a large volume of their potential ad traffic blocked.

This sounds completely reasonable to me, and I don't know where it's wrong, but it doesn't square with my experience and I think it's wrong.

If this were true, nobody would be getting mad about ad blockers. But both advertisers and publishers (big and small) regularly complain about ad blockers, and even with a relatively small portion of users installing a blocker, multiple journalism sites have started adding paywalls. This paranoia isn't restricted to small sites either -- Google's even listed ad blockers as a potential revenue problem in investor disclosures.

A couple of potential causes: maybe it's that impressions actually do matter to revenue in some cases and we haven't completely moved over to conversions yet, or maybe it's that advertising increases consumer spending even without direct conversion rates, or that most publishers are engaged in fraud and ad-blockers make it harder to do that. It might even just be that publishers/advertisers are all stupid and haven't realized that ad blockers don't matter, but I think that's unlikely.

What seems plausible to me is that most of the people who block ads are not as immune to advertising as they claim. If they didn't block ads, they would occasionally click on an ad and provide a conversion. I would not be surprised to see that exposure to advertising increased overall spending, and didn't just redirect it to specific companies. I could be wrong though, maybe it's something else. I'd be interested to see research on that kind of thing.

Regardless, it is very difficult for me to square the world you describe with a world where a huge fraction of news sites I visit are including banners that tell people to whitelist ads, or anti-adblocking technology outright, or are just giving up and switching to subscription models. I can't think of an explanation for that other than, "ad blockers decrease revenue", and I think that having an explanation for that behavior is a prerequisite to making a believable claim that ad blockers aren't hurting revenue right now.

I mean, Google's not dumb. They would not have wasted time on the acceptable ads initiative if they didn't feel like they needed to.

Publishers definitely get hurt hardest because volume and quality of traffic is more important for them, less so for exchanges. They're the ones who are the "huge fraction of news sites [you] visit [that are] including banners that tell people to whitelist ads, or anti-adblocking technology outright".

> maybe it's that impressions actually do matter to revenue in some cases and we haven't completely moved over to conversions yet, or maybe it's that advertising increases consumer spending even without direct conversion rates,

I mean you're right. Branding campaign spend just wants to eat up impressions and don't care much for click or actions. They're a large amount of $$ spent, but they also benefit the least from tracking so they can basically be shown to anyone. Tons of available inventory for them. But there's tons of money in more targeted CPA, CPC ads too.

Disclaimer: I mostly knew things from the RTB angle maybe 3 years ago. I may be out of date with where things are at.

I always thought that the big problem for advertisers was who blocks ads. Adwords stopped being effective for me years ago, looking back I assume that's because I sell to a technical audience who are most likely to run adblockers.

I wonder if that applies to income as well, are people with higher purchasing power more likely to block ads. My gut feeling is yes.

Advertisers converge toward pay per conversion on average, so removing non-converting traffic doesn't affect spend or payouts.

It’s unlike that add blockers are exclusively used by people who are completely unaffected by advertising.

Some banners ads are not all of advertising. You’re affected by numerous channels and mediums that would pick up the slack (and already have) from adblocking on the web.

Does that matter?

If advertising moves off of the web onto other channels and mediums, then necessarily the economics of the web would still change, right? Taking revenue away from a single medium will still alter the economics of that medium.

Mozilla isn't trying to change the economics of the entire market, they're trying to change the economics of specifically web content.

Sure, if that actually happened, but the comment was that ad blocking could do it.

In reality, ad-blocking is already priced in and slowly being recovered. All it did was create a new addressable segment of tech-savvy/ad-resistant consumers. The slack is now recovered in first-party walled-garden platforms, native advertising, "influencer" and other underground techniques.

Or make the demographic info useless by intentionally clicking on ads at random costing them money.

Adding to this, in the same way that privacy specialists struggle to convince the public of the supreme importance of security and privacy, it's difficult for superusers to appreciate the general public's feelings about the trade-offs of advertising—in other words, the general public might never care.

(Mozilla's campaign is a form of activism by superusers, for superusers.)

You are assuming that any effort will be trying to break the advertising paradigm in favour of a direct payment one.

The article/grant doesn't seem to be about that. Whatever the business model, this seems to be more about breaking platform capture than breaking business model. The alphabook platforms particularly.

Ultimately it's a big field. We have direct payment models (netflix, spotify, iTunes, kindle...), ad-suported models and quite a lot of online content production where monetisation is secondary to distribution.

I think it's a good time for this initiative. YouTubers (for example) with millions viewers can (a) make very little money (b) chaffe at various policies, either money or content related and (c) generally have far less power vis-a-vis alphabook than their audience size implies.

Meanwhile, podcasts, a much freer medium, is both more profitable and quality oriented because of the platform openness.

...open doesn't mean no advertising. The main thing it means is decentralised, or centered around content creators.

This a a good idea. YouTube is no longer necessary. If it suddenly disappeared, online video content would quickly recover fully around alternatives. They contribute the least value while extracting the most.

> Meanwhile, podcasts, a much freer medium, is both more profitable and quality oriented because of the platform openness.

I don't think this is true. Podcasts are more profitable since their production costs are lower, their audience skews wealthy and its inherently easier to make longer audio content, so it's possible to put more ads in.

Nothing is stopping YouTubers from not using the default ads and getting their own sponsors.

The web was fine without these commercial companies. Not like their content added a ton of value to the web. The news is mostly propaganda anyhow, belonging to various wealthy parties that want people to pay for their propaganda... If they funded it completely out of pocket, that would be a more honest approach to these "businesses".

The other companies like FB, they took bunch of content that belonged to users that users had tools to publish for free and willingly put them in silos.

There is maybe 1% of content that needs saving. The rest of it is the same sort of drivel that you get from tabloids and I say making that stuff economic non-viable is a good thing. The elites though care about making it work because otherwise the democratic media is "fake news", meaning news heavily slanted to what people want to believe rather than what they are told to believe. If you want that, then they should pay for it. Not the people through taxes as Bernie is proposing... or this scheme.

Other than that we have good models like BBC or PBS on how to do independent government and publically funded programming.

In fact if we re-engineered the web so that there was no data sharing between various systems on the frontend, something like everything has to be served from the same canonical domain, and put it in law that you could not share data between various private companies on the backend without heavy regulation and oversight that would be a better web. I mean prism is already sharing all this data, why don't we put controls on that?

The web was hands-down better when most content was made by hobbyists without a profit motive. I used to be able to spend countless hours online. Now it's rare for me to stumble across anything worth looking at. Granted, it's easier to search for and find specific information.

Totally, just like so many bands were better young, clubs, beaches and restaurants better before they got popular, etc. Unfortunately it's just something that happens when things get mainstream, you can't reverse the process.

You mean like censorship? Who gets to decide? Or do we continue to upvote like we do now?

Not sure what you are talking about, can you quote what you mean? No one gets to decide, I mean personal information being shared between companies on the backend. That should be heavily monitored with oversight. Like the sites you visited, etc.

> [all consumers' discretionary income they'd be willing to spend] is likely a small fraction of the size of [all commercial companies' marketing budgets]

Assuming you meant all spending, this cannot be true, because all marketing budgets are paid out of people's spending. You are paying for advertisements, they are included in prices of goods and services you buy.

I can't forego buying food and housing and medication so that I can reallocate "what I pay for advertisements" to tipping websites. If there's less advertising inventory available on websites, P&G is going to reallocate that spend to some other form of advertising, they're not going to drop the price of the shampoo and toilet paper I buy from them. The money's not in my pocket to spend, it's in theirs.

You can reallocate. Only top tier brands use ads, and the price of the advertising is part of the premium you pay for the brand name. For a lot of products, there's no functional difference with their cheaper alternatives.

So, this brand premium is in fact discretionary spending that you could reallocate.

This is kind of an academic point. "Reshaping the economics of the web" probably shouldn't involve assuming all consumers will want to change what brand of laundry detergent they buy so they can read more blogs.

Why not ? Amazon has launched an offer where you can pay to access all kind of otherwise paywall gated media content (Amazon video and music and twitch prime etc) while also having 2 day delivery on a set of retail products not necessarily including your prefered brand. And people subscribe to Amazon, which finances the said media content, all the while changing their laundry detergent of choice.

I don't say this must be a direct rational choice, but it's already kind of happening in some places.

It can be true if people spend more than they think they are willing to spend on content via advertisements. That is, people underestimate how much advertisements affect them. And of course there are people who simply block advertisements.

Either that, or advertisers overestimate the impact of online advertising.

marketing pays itself to an extent through increased revenue and scale. Historically, companies have spent more or less the same percentage of their revenue on marketing. It seems they find an equilibrium

> the pool of [all consumers' discretionary income] is likely a small fraction of the pool of [all commercial companies' marketing budgets].

"Discretionary income" is a short-term idea, defined as [income - cost to maintain current lifestyle]. Over longer periods, lifestyles expand to fit current income and the idea is meaningless.

For instance, when cars were new, people could only buy them if they had a large chunk of discretionary income to spend on them. But now, most middle-class people own a car that costs more than their discretionary income. So discretionary income isn't a cap on what people can spend in the long term.

Discretionary income is measured in dollars/time, cars are valued in dollars. That’s relevant because cars are affordable to middle class Americans because of _financing_ (also because of mass production). Your car payment deducts from your discretionary spending while web advertising does not—without web advertising you would need to subscribe to the websites and services you consume and that would come from your discretionary dollars. Whatever your views on the matter, getting rid of advertising isn’t going to come for free.

That may be true. But I have no clue where those marketing expenditures go. I do recall reading about major firms realizing that most of their marketing expenditures more or less accomplished nothing.

From a bottom-up perspective, I've read that ad income is typically on the order of $0.01-$0.10 per page view. As a more or less middle-class American, I'd be willing to pay that. Or at least, I would if ad blockers stopped working.

And while that wouldn't be affordable for most people worldwide, I'm sure that there's some way to normalize by user geolocation. In fact, I wouldn't be surprised if ad payments aren't already normalized by user geolocation.

My thoughts as well. Between GDPR making current ad models harder to pull off and more risky and the fact that ads are getting less effective I think I might see an advertising bubble:

Here are my thoughts, comments on why this is wrong are welcome: the market for online ads seems to have been growing more than consumers disposable income for two decades.

Much of that might have come from cannibalizing other forms of advertising, but sooner or later it can't grow anymore as a new equilibrium has been reached between TV ads and online ads and consumers don't have more money to spend anyway.

I remember an old blog post from Google or something showing a whiteboard with crazy thoughts: what if we could have a mirror behind the earth to reflect sunlight and make the day longer? Then people could spend more time searching => more ad impressions => more revenue.

I feel we are at that point now, only consumers have enough time to browse, they just can't buy more than they already do (and some markets might be even beyond that point, fueled by credit card debt etc.)

I've argued this point before, and have hoped for comments from people who run ad-supported blogs. Or manage larger commercial sites. Just to see if my estimate low-balls ad income. But no ...

I do believe that we're in an ad bubble. For one thing, I've read that it's hard to measure effectiveness. You can look at revenue vs ad spending. But so many other factors affect revenue. Some you have measures for, such as how the economy's doing.

But for others there's no public data. Such as ad spending by competitors. You can measure that directly, or buy data from firms that do. I gather that's a key use for VPN services with residential IP blocks.

And then there's the fact that you're typically running multiple ad campaigns. So it may not be obvious which of them actually increased revenue.

Bottom line, I suspect that firms have been throwing money at online advertising. Either blindly, or seduced by bullshit from ad management firms. And so it's entirely possible that total ad spending is greater than users' discretionary income.

But if that disappeared, and was replaced by user micropayments, I doubt that all sites would be impacted equally. HN and Wilders wouldn't be impacted at all, for example, because they don't run "ads" per se. And popular sites generally could probably earn as much as they do now from ads. At least, if their paywalls couldn't be gamed.

So is there anything that would disappear that's valuable? I can't imagine that anyone would miss ad-supported link farms, for example.

> the fact that the pool of [all consumers' discretionary income they'd be willing to spend] is likely a small fraction of the size of [all commercial companies' marketing budgets].

I don't understand this statement. The companies marketing to people with discretionary income presumably get their revenues from that discretionary income. So mathematically their marketing budget would have to be smaller than the revenue from these people.

> presumably get their revenues from that discretionary income

This is where you made a mistake. Their revenues comes from spending on necessities, not from discretionary income. Proctor & Gamble's $7 billion a year in advertising is coming from our buying soap and laundry detergent and toilet paper, not what you have left over for movie tickets. The $30 billion a year in medical marketing is coming from insulin and blood pressure pills, not your Netflix subscription. Most consumer spending is on necessities (food, housing, car payment, health/auto/home insurance, mortgage interest, fuel, etc) and those companies do the bulk of the advertising. People cannot choose to spend that money tipping websites instead unless they stop eating, stop taking medication, etc.

I don't think that distinction is precise enough for this. Sure soap is a necessity, but if you buy some premium fancy soap from P&G that had a lot of marketing budget contributing to its high cost rather than the supermarket brand plain soap, then plenty of that soap money was discretionary.

This extra precision doesn't change anything. None of the proposed alternative monetization schemes have suggested you buy a cheaper brand of soap so that you can put another nickel of discretionary income into tipping websites. Even some of the money you spend on the generic store brand ends up in someone's marketing budget.

By [all consumers' discretionary income they'd be willing to spend] I think the parent meant [all consumers' discretionary income they'd be willing to spend on web content] which is obviously a subset of all their discretionary income. Marketers are peddling tangible useful items, and people will pay a lot more for that than for the ability to read given words on the internet.

That first sentence took me a little while to wrap my head around and I don't think it's true or logical. In other words, even if it was true, I don't see why it matters.

I'm building a company that plans to charge people to read things. The reading experience, without ads, is superior. It's worth paying for. That means that an ad free future is possible.

You may think it's worth paying for, but your average consume will not. It makes sense from a consumer point of view, how am I to know if this "superior" experience is worth my money if I haven't tried it. How do I know it will stay that way, lots of magazines and newspapers quality tanked after they got high subscription numbers.

"Did you see that new Thai joint opened up down the street? I heard it was pretty good. Want to try it out?"

"And just how are we to know for _sure_ that this trumped up restaurant is truly superior?"

In other words, this is how capitalism works. Businesses and vendors make value propositions, then people take chances on them. It works for every single other class of good, but for some reason the big brains that run the internet can't wrap their head around how it would work for them.

Some people have understood this, and they're making a killing from it. For instance, which do you watch more of, HBO, Hulu, Prime and Netflix or broadcast TV? Tons and tons of people are happy to shell out cash for products they consider worth it.

Newspapers and magazines have tried to convince us that these asshole millenials refuse to pay for things anymore, but what they mean is they refuse to pay for shitty things. They're actually perfectly happy to OVERpay for things they like. It just turns out newspapers and magazines aren't, you know, worth that much.

Forget the average consumer...it's pretty much guaranteed there will be an outline.com link posted on any paywalled article on HN (along with the usual paywall complaint) and this crowd seems to fancy itself above the average consumer.

>It's worth paying for.

Given the current state of the web and the struggle that many media companies are having, I don't think we can take this for granted today.

Fingers crossed that you and experiences like Mozilla's proposed can change this.

Maybe your business attracts customers now because they have a lot of discretionary income to spend on it, but if other services that are a higher priority to your users start to find a subscription model to be more valuable than an ad model (because ad blocking is too effective, for example), your customers will have less discretionary income to put to ad-free reading. Further, all of those ad-subsidized reading sites now become your competition.

This isn’t a pro-ads argument, but an appeal to sober, informed debate.

I think you’re right, there is definitely room for consumer-funded content in the market - and I wish you the best in your endeavors.

But I think suggesting an ad-free future is a leap, not least because paid content platforms need to do marketing. Why do you have a Netflix subscription? Because you heard they had some good shows? how do you know what shows they have before you subscribe?

Does it say somewhere it is tip based?

It seems like there are a ton of things that can be done to make it work better than where it is totally based on "small tips."

That's not exactly how it works.

I don't price my products depending on marketing.

Marketing is a % of my profit I'm willing to spend, to try new things.

That's not how it works for most businesses.

Why does it not adress it? It is also about changing the mindset of people. Make it easier -> make it more common to voluntarily pay for thing you value -> more projects are financed that way -> more people see it is a way to make a living -> better content -> more people willing to pay, especially seeing the difference to normal ad madness

Nothing in that feedback loop will allow people to spend money they don't have. A big chunk of the money you spend on groceries, car payments, insurance, medication, etc is going into those companies' marketing budgets, and that ends up as the ad revenue that finances the web today. You don't have that money to give to website owners directly instead, as you've already spent it on those goods.

Marketing makes people buy plenty of things they don't really need nor want. If people choose not to be exposed to tailored ads so much, they would have this money to spend for other things, like paying directly to support what thes want and not the long route indirectly and ineffectivly via ads.

Most marketing is for necessities, to steer which brand you decide to buy, not to make you buy a category of product you wouldn't buy otherwise. You need soap, they just want it to be Dawn soap. You need laundry detergent, they just want it to be Tide detergent. Etc etc. You're not going to stop buying food, soap, medication, gas, etc so that you can direct that money to tipping websites instead.

I do not agree with your premise. The amounts that companies make of blatantly selling out your data to be abused is often much lower than I would be willing to pay. I pay for YouTube Premium primarily to get rid of the adds. I stopped using Facebook in 2017, but if I used it I would gladly pay the single digit dollars per year they seemed to make of me.

Now where i do agree is that even with people paying for content and services directly, advertizing would creep back in, because it would always be a potential supplementary revenue stream just sitting there tempting the content or service providers. This is where we need strong regulation such as the GDPR enforced to stem this general race to the bottom.

Facebook's ad revenue for 2018 was $55.8 billion from 2.7 billion MAU. That means each user is worth $20.67 per year. In reality, not all users are worth the same, and as an English speaker, you're probably worth at least double the average. So, figure $40 a year to remove ads from Facebook, not single digits. Now what about the other 10,000 websites you touch each month?

About $50/month for a family plan should be fine for a completely add free internet. How that fee is distributed across the different sites I visit is an administrative issue.

If online advertising went away overnight

I'd extend that to all advertising and I think we might be out of this mess permanently.

It seems like an insane concept, because advertising is a part of modern life, but I think it's the only way to effectively kill the surveillance/influence peddling hydra.

Otherwise, all these efforts just redirect ad dollars to other forms of systems of influence which may be similarly as invasive.

What counts as advertising?

Everyone can have their own definition, and allow me to present my own.

Advertising is any unsolicited communication about a product/service.

That covers ads on webpages, billboards, TV ads, radio ads, youtube ads, someone yelling at you in the street about their product/service.

I didn't look for that info, so I don't want it. You and your product/service can fuck off.

"Advertising" that I do not object to is where the "customer" actually looked for it. Here are some examples of that: Displaying a list of products/services that you provide on your _own_ website that people can look for. Entries in the Yellow pages or something similar. Like for example (unpaid) search results.

If I am looking for a plumber, then I can look in the Yellow pages and see a list. And I won't get pissed off because it's what I wanted. Or I can google plumbers, and get a list of websites in the search results, and then click on the links and view their websites where they describe their products/services. I also won't get pissed off by this, because I was looking for it and no one shoved it down my throat.

I agree with AndrewKemendo that advertising (as I have described above) is insane and needs to go away. It has no place in a civilized society. There is literally no difference between advertising (as per my definition) and spam. I would go so far as to call it harassment.

This is the same as saying "Define Pornography." You're asking for a bulletproof formula that doesn't exist. That doesn't mean we can't take a pretty good bite out of it.

How about this: Any action taken by an individual or organization with the intent of influencing another individual to execute a commercial transaction.

No billboards, TV commercials, newspaper advertisements, hyperbolic statements about products. I'm sure some wacky new loophole would be found, but it would be just that, some wacky loophole, not a massive trillion dollar industry built around manipulation.

Without a formal definition it's impossible to legislate or enforce.

Take HN - arguably this whole forum is an ad for YC. Would this fall under your definition?

> Any action taken by an individual or organization with the intent of influencing another individual to execute a commercial transaction.

That definition pretty much boils down to every interaction between businesses and individuals, ever. Any kind of negotiating or sales or communication..

Correct. Like I said, it's pervasive to the extent that we just assume it's the only way it could be, but it's behind every terrible thing.

I know this first hand, I built these systems. The end state is to know and predict everything you do, and then sell you something about it. Whether its a product or an experience or a feeling. That's where it all leads - you lose your agency. Getting rid of online marketing or nudging it one direction or the other won't really change it. You can ban it online outright, but it will persist in another form.

Our entire world economy is based on different groups vying for your attention to monetize, to the point that as we go further, nothing is not monetizable.

It's the Truman show, but everyone is Truman.

I've worked with small businesses - the small one to thirty person plummer or photographer or decorator type. What you're asking for is fundamentally impossible. Business and entrepreneurship is a human process where two entities find a transaction that benefits them both.

Of course this breaks down when shady actors manipulate in bad faith, but that's the price of freedom. To put a ban into place as extensive as you're suggesting would be to regulate speech as far as I'm concerned.

You're just reinforcing the point that we're in too deep for it to be a comfortable transition.

Not really. I understand your complaint, but what you're saying would kill entrepreneurship. How else can companies compete if they're not even allowed to go 'hey, I think we've got a better solution at a lower price'.

It's crazy, that we still do not have something like "Cash" on the internet.

A simple, anonymous way to pay 10 cent or so. To use a website or read an article.

That would make the world so much better for indie developers.

Currently, an indie dev makes orders of magnitude less money per pageview then Google, Facebook and Co. Because those have all that advertising technology and ecosystem that indie devs don't have. If users would pay directly, that would change.

Monero (getmonero.org) has many of the properties of cash on the internet. Definitely more so than Bitcoin. It's surprising how few cryptocurrencies take privacy by default seriously.

I honestly don't know anything about Monero. How does it compare to Stellar, which was recently on HN due to the Keybase airdrop giveaway?

(I hold a whole $40USD in Stellar, I do however think it's a good thing. But I think there's space on the net for more than a couple of cryptocurrencies so I'm curious about Monero)

Forget Monero it can't scale for micro payment. Currently it does like 1 Tx every few seconds. We can assume it could handle a few Tx per seconds. And with proper hardware upgrade it could probably handle some dozen Tx per second. Anyway all of that is far far away for usable if people want to do micro payments to websites they read.

Stellar is way faster the XRPL even more so (1500Tx/s) but even then it can not be used to stream money. You dot want to go to a site and instantly donate 0.001cent you want to donate only if the content is good and the best way to measure that is by time. Hence Coil steams the money as long as you are on the site. That only works trough ILP not direly on ledger (on-chain). IPL can stream money from any kind of wallet. So it would work with Monero as well but its unlikely to gain traction as its key feature (anonymous) is rather useless in this case. and Tx are slow and expensive compared to XLM or XRP.

Thank you for the information. There's obviously a lot more I need to understand regarding crypto, but you helped :)

Monero is the largest privacy-focused cryptocurrency in existence right now, by pretty much every measurable metric. It's also the only (in the top 250 by market cap) cryptocurrency that enforces privacy at the protocol level, by default. You can read more about it at the website below.


Yeah, I'd love to see more websites accept tips in XMR (and see the Monero economy grow overall)


It's perfectly on topic here. I would hardly call this shilling.

> A simple, anonymous way to pay 10 cent or so.

Pretty sure it solves that.

I thought XMR had high transaction fees?

Transaction fees have been a fraction of a penny since the Bulletproofs upgrade in October 2018. https://bitinfocharts.com/comparison/monero-transactionfees....

Only if you consider $0.002-$0.006 per transaction high. The protocol has been significantly streamlined over the years.

I mean just buying something without giving them your email or filling out a ton of fields would remove a ton of friction.

I could see myself paying 25 cents to read a NYT article right now as an impulse buy. But how does it even work right now. They’d probably ask for my email, phone, and address. Then I’d have to enter credit card details.

You have to register at Coil.com You pay with CC or whatever you like to Coil. Your browsers streams money as soon as you visit a site that implements web monetization over IPL. The site doesn't need any info from you it has the address to stream to in the HTML of the site. So your payment is theoretically anonymous ofc the site can still see your IP, fingerprint your browser etc.

Brave is doing exactly this with BAT on Ethereum but automating the whole thing. You can donate to any website, but you'll also auto-contribute if you choose to have that enabled.

I wish brave would just go for a plug-in approach, and partner with some news sites.

Major publishers are already in bed with Apple News, which is preinstalled on iPhones and iPads. Brave is my goto mobile browser, but it's slow on desktop and its market share is practically zero.

Yah been thinking about trying Apple news. Brave browser really isn’t taking off, that’s why plugins for chrome, Firefox, safari would be better.

There's probably a different cryptocurrency for each user on HN, certainly enough for people to argue over until the end of time.

Perhaps it's as simple as this:

1) Make adding a box to pay a website in your crypto easier than adding google analytics or facebook share links.

2) Make buying small amount of your crypto easier than anyone else's crypto.

3) Give lots of your crypto away to as many folks as possible to put lots of it into circulation.

4) Provide easy ways for people to buy amazon and walmart gift certs with your crypto. (so web site operators can do something with the coins they receive)

Has anyone done step 1 yet?

https://unlock-protocol.com ;) [disclaimer: I made this!]

This is really interesting! I was fascinated by the website, joined the Telegram group… and of course got kicked for not sending a message in 60 seconds. Is there any chance I could email you my username and be re-added?

Sorry, our anti spam bot is wayyy too agressive. We removed it. Please join again (or email me julien@unlock-protocol.com). Thanks!

OK, you win :) Nice job!

Thanks ;p

Unfortunately the service that is free will always be more popular and ultimately win over more users. People have decided that paying with their privacy is better for them. I really wish paying 10 cents to use the service was the norm, but alas I doubt it will ever happen.

I suspect you're right that people will strongly prefer free over paid, but it's hard to know how many people would pay ten cents for something, because it's not feasible to charge that little. Paypal has a micropayment rate of 5% + five cents (USD), which means if you charge ten cents, you get maybe $0.0045. Plus, you have to do all the billing infrastructure etc.

The economics would be better with a stored balance system, but those are sketchy --- you'd need the balance holder to be someone you trust already; or a credit system where you pay once you've accrued enough content debt to make billing worthwhile, but that's going to be totally scammed.

"Free" is also a very psychologically special price. Micropayments would make it easier to get over the energy barrier to paying for something, but it's still there. Laboratory experiments (and real-world experience) show that people value the difference between a price of 1 cent and 0 cents much more than the difference between 1 cent and 2 cents.

Which is why spreading £10 over thousands of websites is much easier to sell than £0.01 per article. With the later model people become paralised with choice, and the former model has been successful as the backbone for SMS/data infrastructure almost everywhere.

It would be nice if there were a reciprocal subscription model. All subscribers to the NYT can also view articles of the LA Times or some other site, but reciprocal logins are tracked and funds are divided accordingly. That way, with some limits, each site could set their price, offer the bonus of visiting other sites, and there would not be the huge barrier to entry of requiring a massive player to aggregate things and wield power over publishers they don't want to accept. It's basically the model of reciprocal agreements used by the postal services around the world.

It's also the model of internet peering, but it requires an even exchange otherwise the imbalances lead to failure.

As an example, the current dispute where the US is leaving the Universal Postal Union (a month from today) because the exchange is unbalanced due to e-commerce from/to China.

What about your ISP? You already pay them for Internet access, how about adding a "content charge" to that, then have some neutral organization disburse that cash (based on traffic? usage? evenly?) to individual websites who promise no third-party advertising, no tracking, etc.

> Paypal has a micropayment rate of 5% + five cents (USD), which means if you charge ten cents, you get maybe $0.0045.

You're off by a factor of ten here, it would be 4.5 cents, not half a cent.

I certainly am! Unfortunately, it's too late to edit to show what I meant.

Fine. Let Facebook be an order of magnitude more popular then my website. I would still be happy to have 240 million monthly users who pay me 10 cent.

Why would I care that FB has 2.4 Billion monthly users and squeezes more then 10cent out of them via ads? That's none of my business.

You can already do whatever you want on your own site, like a paywall instead of ads, so what’s the issue?

Netflix&Co. are more expensive than bittorrent, and yet the former have gained significant market-share over the latter over the years.

If you want to compete with free you have to offer superior quality and convenience.

And being legal has absolutely nothing to do with it?

That's part of the convenience, no pesky letters.

I don't understand why we don't pass a Terms-of-Use tax. Require every Terms agreement to be filed electronically with the state and the end-user to pay 10 cents for that service. I think we need to force their business model to be non-free, especially when they are making people agree to things that have a monetary cost either explicitly or externally.

> People have decided that paying with their privacy is better for them

The hell we have, there's just no alternative other than not using what have essentially become utilities at this point. I applaud people for whom not having Facebook and Google and WhatsApp is an option, but it's not for me.

I'd gladly pay a large chunk of cash to have those companies not be doing shady shit with my data and showing me adverts, but that's not an option available to me.

People decided that paying nothing is better for them. I don't believe most people are even aware of the privacy implications or how that ties into the scheme of not paying anything. It's kind of like how most people don't really know the economics of Casinos.

Doesn't mean people don't value privacy. Just means that people don't value online services and content.

That may not be true.

I don’t think it’s been well tested, and we should at least try giving people the benefit if the doubt before we give up, in my opinion.

The answer to this was meant to be Bitcoin, hence the title of the whitepaper "Bitcoin: A peer to peer electronic cash system". Unfortunately there is a lot of money being spent to keep that goal from being realized.

If you really want to see "cash on the internet" I would advise you to do some research and find the cryptocurrency that best resembles it, then help build it. It will be an uphill battle the whole way.

I think there are a lot of good candidates. Personally I am rooting for Ethereum. But Bitcoin Cash, Litecoin, Monero, are all good options. Even Bitcoin could be a good option but it seems like you said there are too many people working against that.

Well that is the whole premise of this grant.

Coil, the co-sponsor of this grant, is a company founded by the ex Ripple CTO and trying to do that.

I think Twitter is missing out on a big opportunity for micropayments by letting users "tip" real money when favoriting tweets. Content creators would be incentivized to use Twitter more if they can make money directly from tweeting. More content from more content creators would also attract more users. Content stealing and reposting would be a bigger problem, but you can somewhat address it with text and image matching algorithms and user reports.

For Twitter there are bots for various cryptocurrencies which you can mention in an answer to a tweet, leading to the original tweeter's wallet being filled. Unfortunately most systems are not decentralized (custodial cryptocurrency wallets) and of course there's no consensus on using a single cryptocurrency.

> micropayments by letting users "tip" real money when favoriting tweets. Content creators would be incentivized to use Twitter more if they can make money directly from tweeting.

I'd honestly delete my Twitter account if they introduced that. Twitter as a social media channel is fine (to keep up with creator's content elsewhere and maybe the odd follow-up generated by that content) but there is zero content in my feed I'd be prepared to pay any money for (and that would still hold if it was longer form/mixed media).

There's been dozens of these products. Some have even been supported by Google, Facebook, and Co.

Nobody uses them.

Can you give a few examples?


The gp post is correct. When given the option between an ad, adblock, or tipping, people overwhelmingly choose option #1 or #2.

They then complain about ads, if they chose #1, and about adblock naggers, if they chose #2.

And, of course, all three sets of people lose their mind when they run into a paywall.

There is no non-legislative solution to this problem.

Google contributor most likely was a failure because they didn't do anything about promoting it. I only learned about it on hn and after it was already dead.

Simple, paywall but provide an option to pay a reasonable price for an article, say $0.25 instead of tipping or ads. Of course NYT and all want a sleezy $5 subscription that’ll balloon to $30/mo in three months.

I think indie developers shouldn't rely on advertising at all. It's like being adventurer and rely on carrots as food for adventure. You need to find the proper food to do adventure. Same with the developers. Indie developers must do software and provide a service or anything that is valued by a consumer. Anything else is just a waste of time.

Having ads doesn’t mean the context has no value. Otherwise why are you visiting any site with ads?

Indeed. But ads shouldn't be the reason you write software or write about software development.

Who’s doing that? And nobody needs to tell anyone else why they should or shouldn’t do something. Everyone can make their own way and living.

Sure. I never put that power of choice. I was just stating my opinion. The market will regulate everything without us stopping others to keep doing things for ads.

GNU Taler is a WIP for that.


>Taler is a pure payment system, not a new crypto-currency. As such, it operates in a traditional banking context. In particular, this means that in order to receive funds via Taler, the merchant must have a regular bank account, and payments can be executed in ordinary currencies such as USD or EUR.

That implies high transaction costs to say the least.

Very interesting, thanks for posting.

> Currently, an indie dev makes orders of magnitude less money per pageview then Google, Facebook and Co. Because those have all that advertising technology and ecosystem that indie devs don't have.

If you're referring to the cut AdSense takes out of ad revenue, it's about 30%

Simple? Anonymous? Sounds like a perfect money laundering scheme.

All governments in the world hate cash in hands of citizens. It's hard to track. It can be used for tax evasion, buying illegal stuff, financing of undesirables, any transaction that a government would not approve of, but cannot easily learn about.

Many countries evict cash from daily use, either leaving it for small-scale purchases ("we only accept notes of $20 and below"), or even completely.

The chance of an anonymous online currency to appear is nonzero. The chance of it to be widespread, convenient, and not under constant legal scrutiny is epsilon, for it to last, epsilon squared.

My idea was just to have this integrated into your ISP bill.

I once thought that ISPs should pay for all content, then pass the bill onto users + costs-and-profit. I guess there would need to be separate standardized rates for video, audio, and text, but that creates a ton of regulatory and technical issues. Not the least of which is people don't want to pay. It would be like music licensing essentially, and look how that played out.

So just make it all voluntary like patreon, but integrated into my ISP billing, and give me an app/browser add-on to make payments and track my spending.

Say there are three people living together in a share house/unit, all sharing the one internet connection. There'd have to be a way to deal with cases like this.

Security issues. Ad fraud is already bad enough.

And small children paying a thousand dollars for an app does reach the news.

I started working on this exactly 2 weeks ago, so was pretty surprised to see this post by mozilla just now!

Essentially a very simple browser extension to tip websites, where you load some credit and it gets distributed as you browse (with more nuances obviously, but that is the main idea)

Even if it never amounts to anything it might at least inspire others to start similar projects until something better than ads appears!

Even though it isn't without controversy, Brave browser has this exact feature through use of its BAT token [1].

Basically, you earn BAT either by directly depositing money into your wallet, or by viewing "privacy respecting ads" (herein lies part of the controversy), which let you earn 70% of the ad revenue as BAT. These tokens are then distributed over the content creators who have signed up for the program. The exact distribution depends on how much attention you gave each creator.

[1] https://brave.com/brave-rewards/

It exists: http://opentransactions.org/wiki/Main_Page

It is, like all crypto currencies, bottlenecked by the issue that it has to be exchanged for hard currency.

Cryptocurrency can do this, but Bitcoin artificially keeps the blocksize to 1.6KB/s so people think that small transactions are not practical in a general sense, since they aren't practical there.

The throughoutput is only one of the problems. There are more:

1: You need to download and install software to use current cryptocurrencies. And users hate that.

2: Turning Fiat into Bitcoin and vica versa is not easy.

3: It is less private then cash.

4: Volatility of the value. Users would want something pegged to their currency of choice.

> You need to download and install software to use current cryptocurrencies. And users hate that.

People already trust companies with their money. If banks supported crypto natively the no-install/web app model works great. The last three points are rendered irrelevant here if banks are able to make the experience easy and guarantee some value.

Those who don't want the protection of a bank are trading off points 1/2/4 for point 3.

> You need to download and install software to use current cryptocurrencies. And users hate that.

Users don't have a problem installing apps on their phone.

The rest are valid problems, with #2 and #4 being directly linked to low adoption.

These aren't show stopping problems.

Web pages and browser extensions are pretty low barrier to entry.

Exchanging cost is not too bad, but probably the biggest hurdle still.

Privacy I don't think is a problem here. One random address to another is good enough for these type of transactions.

Pegging to another currency means centralization, though that is acceptable to plenty of people.

You said you wished there was a solution, all I'm saying is there is a clear path to how it could be done.

It wouldn't be the best user experience if you had to wait 10 minutes for confirmation. Also by increasing the size of the blocks you increase the orphan rate due to propagation delay which results in needing more confirmations before a transaction is considered truly immutable. Bitcoin has to have a fee market to incentivize miners so that as block rewards decrease the cost of securing the network doesn't shrink as well.

You don't have to wait 10 minutes for a confirmation, especially not if we're talking about dollar amounts. Credit card transactions work exactly the same as 0-conf, and we seem to accept them just fine.

> Also by increasing the size of the blocks you increase the orphan rate due to propagation delay which results in needing more confirmations before a transaction is considered truly immutable.

Which happens at around 20 MB. Improvements like Graphene, Xthinner and blocktorrent will drastically reduce the orphan rate allowing for much bigger blocks to be propagated fast enough for the orphan rates to not climb in a relevant fashion.

> Bitcoin has to have a fee market to incentivize miners so that as block rewards decrease the cost of securing the network doesn't shrink as well.

Or Bitcoin needs many more lower fee transactions to compensate. In a couple of decades.

Bitcoin is never going to solve anyone's problems now. The people who have taken it over have relegated it to a lesson in what not to do. Shorter block times are not a problem and neither are much larger block sizes, both of these things have been proven by other currencies.

I've never had an issue sending small amounts using Bitcoin Cash which has increased the blocksize limit. The guy who took over after Satoshi left also says that Bitcoin Cash is what he was working on for what it's worth. https://twitter.com/gavinandresen/status/929377620000681984?...

Cash Shuffle and Cash Fusion will also provide the privacy needed for users to safely enjoy Bitcoin Cash.

I'm not going to pay for every stupid little thing I do on the internet. And I'm sure there have been attempts at stuff like this, but who really wants this?

I can see it now.. every link I click on here or reddit I have to pay $0.05 to read. Yeah, no thanks. Suddenly the most upvoted things will be ones that people promote to make money on etc.

I already have to pay for every damn thing when I leave my house.. when I'm at my house browsing the internet I would really like to not have to constantly be mindfucked with small purchases every moment.

> when I'm at my house browsing the internet I would really like to not have to constantly be mindfucked with small purchases every moment.

Do you feel this way about using the toaster, shower, every light switch, turning on the TV, marginal usage of rechargeable electronics, etc? You already make "micropayments" for a million tiny, less-discrete usages of electricity, but have learned to adapt to only getting feedback and adjusting usage at a monthly cadence. Why do you assume this couldn't also be the case for Internet usage?

I don't know if I would mind so much donating, say, $20 / month worth of microtransactions to content I really enjoy...

The problem would be the friction of deciding what I reward and what I don't. I don't want to make 200 decisions like that every month in my day to day browsing.

If you bundle it, then it's basically just a subscription (I already pay for, say, the Atlantic).

If you let an algorithm decide for you, then it probably becomes a shitshow.

Maybe there's a genius idea to solve this massive UX problem but I certainly don't see it.

>The problem would be the friction of deciding what I reward and what I don't. I don't want to make 200 decisions like that every month in my day to day browsing.

Yeah that is exactly the problem. It would be like having an ala-carte TV cable where you were charged every time you turned on the TV to watch a certain show. I just wouldn't do it.

Yes, if only the powers that be could let go of dictatorial control of our finances.

But that's just not the way it is. Nothing anonymous can be tolerated.

This has been attempted so many times and they all end in obscurity of failure.

Might be about timing. Maybe this time it could be different.

Halfway through reading this, the site suddenly covered half the screen with a "donate to Mozilla" banner. The case for "reshaping the economics of the web" would have been a bit more compelling without that...

(My problem isn't about asking for donations. It's about asking for them in a particularly user-hostile way. If the call to action at the bottom isn't enough, work it into the text.)

Super ironic too, considering they are working on blocking in-page JS popups like this and there's an addon to report them. https://github.com/ehsan/popup-reporter

Also ironic asking for donations on a post about giving away $100,000,000.

The funding comes from Coil.

I find this weird. I believe most people who donate to mozilla want to do it to keep the lights on for the browser development.

Now they are simultaneously asking for donations and giving 100 mil away. Non profits donating money to other causes seems wrong to me, I think it would be better if they offered an option Humble Bundle-style to split the money and let the donors choose

> Funded and led by Coil

Its sort of fitting though. Obtrusive popups like that are prevalent because, unfortunately, they work better than anything else. If something less obtrusive worked, then this particular grant would not be needed.

Essentially Mozilla is offering people money to find a way to make that banner unnecessary.

I instantly put a permanent block on any site that does this, and never visit it again. So that's Mozilla done for me.

I dunno, I could see that as being the most effective thing possible: I mean, if Mozilla wasn't feeling the need to do that, the problem might already have been solved.

There's nothing new here. Micropayments have been tried countless times. We did it ourselves and came to the same conclusions that people just don't value content highly nor want to pay for it at any sustainable rate. [1]

Advertising is faster, easier, more passive, and more egalitarian than direct payments. Many people you might not consider (like billions around the world) are happy to view ads in exchange for free content and services they couldn't otherwise afford. Also payments just means you work, earn cash, then pay. Ads are on-demand cash generation that uses your attention in real-time to pay exactly when and where you need it.

Privacy is also much more nuanced than these extremes. Most people are clearly comfortable sharing a lot about themselves on social media. Transparency and control is a far more important and productive goal than fighting over whether anyone or noone can use data.

1. https://news.ycombinator.com/item?id=19038820

It sounds like Sterling was some cool work - too bad it didn't work out. If it's going to happen, I think it's going to require a non-iterative step-change to get it right.

A far cleaner design for the web would be essentially an underlying payment system that governs all web traffic. You would essentially be paying for bytes, at a fairly low level. The issue is, people want to be completely unaware of the payment process so that their browsing habits are not perturbed. And yet at the same time, they must be able to trust that they won't be hit with a gotcha charge from some rent-collector troll site. You can't have both without some level of oversight, or some seriously negative consequences for people who set their "tolls" too high.

One of the biggest advantages of pay-to-play is deterrence of astroturfing and spam. If it costs money to send an email, the losers are the parasites who send millions of unwanted emails at everybody else's expense. But for these types of second-order social effects to take root, you really need a wide rollout.

The modern-day web is exactly backwards - it charges bots nothing and real humans everything. When a website is starved for bandwidth because demand for the content is higher than the web's ability to supply that content, it just breaks. It becomes the host's problem. In any other arena, the prices go up, and demand backs off until supply returns. Having tons of people demand the content you're serving from your website should be a _good_ thing, but right now it means you need to scale or die and cut off your revenue stream completely.

If it were profitable to be able to serve web content to people, then the decentralized web would become an obvious and inevitable reality, especially over the final mile. Rather than having to sign up for Comcast, they would make their money invisibly, behind the scenes, through the magical internet currency that was used to pay for bytes. You'd never have to talk to some fuck about your xfinity account again. No website would ever have to think about how to pivot into profitability. No youtube streamer would have to shill for products in the middle of their videos - the fact that you're watching it would be making youtube money directly, which youtube would pass on to the creators at a far more transparent way. No more demonetization because some companies might not want to be associated with you. Everything would be better. Everything. Making the web free is actually part of what led to its zombification.

A big problem with micro payments is that the effort that goes in to figuring out if that article is really worth the $0.27 to read it is often worth more than $0.27


Why wouldn't these two approached work in parallel ? An ad-free experience for those willing to pay and an ad supported one for those who don't.

I'm sure there are nuances like the politics you mention in your linked comment. Just curious why.

They can, and some services offer both. Maintaining two systems is resource and support intensive though, especially for payments. It's rarely worth the hassle if you only get 1 in a million to subscribe directly.

The cable-bundle approach to content might get more traction but there are too many competing bundles adding fatigue (just like video streaming), on top of the fact that people don't value general content the same as music or movies so the rates are inconsequential compared to ads.

Companies don't want to advertise to the people with no money.

That's an astute observation, however I doubt it's valid. Not paying a monthly subscription wouldn't indicate poverty. It might just be these people are subscription averse... or they are used to ad supported free content and don't want to pay...

> are happy to view ads in exchange for free content and services they couldn't otherwise afford

So advertising doesn’t work on these people? Seems unlikely. That would imply content providers are giving stuff away, and wasting bandwidth on adds they know will never work.

Seems more likely to me that the advertising does work, and people aren’t getting content for free. They’re paying for stuff they don’t need, with money they don’t have.

No, free in this context meaning they aren't paying upfront with cash. So yes, ads do work, and they allow people without other means to monetize their attention in real-time.

They also own their attention and are perfectly capable of spending it however they want, on whatever they want, without any approval of what they need by you or anyone else.

How did you draw that conclusion?

Forget micro-payments. Ban web advertising at the browser level. Create a browser that thoroughly and intentionally integrates ad blocking and has a statements of how it renders HTML with UX constraints (think Google AMP but actually good).

What happens?

Most content sources dry-up and stop. I no longer see that as a bad thing.

Out every every 100 content mills that are re-aggregating and summarizing someone else's original reporting you can wipe off 90 of them and still be fine.

Those that have a foot-hold, brand, or better reach survive through affiliate marketing (see WireCutter), donations, or subscriptions (NYT, WaPo, etc.)

I'm not saying there won't be casualties in this approach but you lose the alt-right and other reality-bending publishers pretty quickly.

> What happens?

What happens if you do that without enough market share is that the browser gets detected and banned/redirected with a message about it being hostile to the people making the content. To some degree, they would be correct.

If chrome, Firefox and Safari all did that all at once, many web properties would go under before the dust settled. It would probably cause a recession with how many people it affected, if not an outright depression, as it reverberated throughout the economy.

There's a reason some things happen slowly, and why puahback against the better future isn't always a bad thing. Too much change all at once is far more destructive overall than if spread out over time (for the exact same reason having two cars need major repairs the same month can put a family in dire straits).

All very good points and you could have a detailed discussion on many of them.

I still think that this is a situation where the bandaid/knife needs to be pulled off quickly. Chrome will never go that path voluntarily but you could push Safari and Firefox to do so (unlikely).

I've worked with publishers big and small and I know where this would hurt the most with smaller sites shuttering.

That said, after working in advertising for over a decade its clear to me that it is manipulative, abused, and ultimately harmful and directly/indirectly funding several hateful voices that have influence on the world.

Don't forget that it would also make bezos even richer as amazon.com is the only place people know off the top of their heads to buy anything

I dont really know how to change the economics of the web without changing the laws around financial transactions. The various anti money laundering laws and anti terrorism laws make it basically impossible to transact money without significant compliance costs. And if the transactions are to be small and decentralized those compliance costs become prohibitive.

The current answer is to bundle. Use a subscription for content that is then split across sites/content by time/attention/some other measure.

However the problem then becomes which bundles you subscribe to, and that can easily cause fatigue and a race to the bottom as seen by all the current video streaming services now. Even worse, none of it has proven completely profitable or sustainable either.

I used to use flattr and thought the concept was sound. Until I saw that most of the creators I wanted to sponsor never bothered to sign up and collect their money.

Anyone else have experience with flattr or similar alternatives?

I figure with all the "block chain" interest we'd have seen something built on a cryptocurrency allowing "direct" microdonations to a website's "wallet"

EDIT: looks like coil is kinda of like flattr

Flattr is the property of Eyeo GmbH, the same company that owns Adblock Plus. Also keep in mind that you will have to give them a copy of your government ID to withdraw the money you receive.

Readability tried something like this. Readers liked it a lot more than publishers did. https://web.archive.org/web/20120702145822/http://blog.reada... https://news.ycombinator.com/item?id=4105891

this comment was telling:

Keith Calder June 13, 2012 • 10:41 am

I considered signing up for the Readability Publisher Program. Then I read the Terms of Service (amusingly I just read it yesterday, so this is all fresh in my mind). There is no way in hell I could agree to those terms. It basically gives Readability the right to do whatever they want with my content (which they were already doing, even without my permission) and I would have to limit any liability Readability might have for doing whatever they want with my content.

Some gems from the Readability Publisher Program Terms of Service.

You shall not “modify, translate, or otherwise create derivative works of any part of the Readability Service”

You know, the exact thing Readability does with my content. Or are we pretending that Readlists aren’t a derivative work?

“You agree to hold Readability, its directors, officers and employees, harmless, including costs and attorneys’ fees, from any claim of copyright infringement or intellectual property misappropriation related to ANY WEB pages or articles sent THROUGH or processed through the Readability Service, the Readability website (readability.com) or through the Readability API.”

So basically if I sign up to collect money from Readability, I have to hold them harmless for ANY copyright infringement on ANY web page or article I have authored that is sent through the Readability Service. So if I wrote an article that is behind a pay-wall, and someone figures out a way to add that to a ReadList… Oh well. Nothing I can do about that, unless Readability decides to be nice to me and remove the article. But they’re under no obligation to do that. They could just choose to let people freely redistribute my content against my wishes because I signed up for their Publisher Program to collect money they decided to collect on my behalf.

And just to add insult to injury…

“The Readability Service are protected by copyright, trademark, and other laws of both the United States and foreign countries. Nothing in the Terms gives you a right to use the Readability name or any of the Readability trademarks, logos, domain names, and other distinctive brand features.”

Nice to know that Readability is protected by copyright and trademark laws that apparently don’t protect any of my content.

So yeah. Good riddance to the Readability “publisher payment plan.” I look forward to the other exciting ways Readability will help great content flourish online.

Remember when people thought that selling unprotected mp3 files is crazy, and everyone woulDD just pirate them?

Well, yes, some do, and will do.

But in most cases buying is much less hassle, and costs very little, so there is no point to pirate just from the convenience point of view.

We need Spotify and Bandcamp for texts, with fast and frictionless payment, cheaply operated.

Satoshipay[0] has been around for a long time (and many iterations) and is now based on Stellar and from what I've seen pretty simple to integrate for websites.

[0]: https://satoshipay.io

I liked Coinhive but despised that 99% used it for malevolent things. I do like the idea of letting my computer do some proof of work for a few seconds to unlock an article. It's like doing a micro payment with your electricity bill, anonymously and without any setup.

In any world in which work that a browser can do in a few seconds will make any appreciable amount of money, someone can make more money by throwing a server or GPU or dedicated hardware at it, and the browser's few seconds of work stops having value.

Coinhive mined Monero, which is ASIC resistant and where GPUs are only twice as efficient than CPUs. It's suited for browser mining. The browser shot the calculated hashes into a pool so it's not really about one browser vs. dedicated hardware. It's about a large quantity of browsers doing the work in parallel.

Suppose the minimum amount to make this make sense is worth 1 cent over 4 seconds of running the proof of work. Then this proof of work would earn 24 hours/day * 60 minutes/hour * 60 seconds/minute / 4 cents/second / 100 cents/dollar = $216/day. But any proof of work system that can earn $216/day would make far more sense to run dedicated miners for instead of running miscellaneous computers on.

Here I thought Mozilla had made it possible to do micro transactions in the browser. That Mozilla had started a bank backed by 100 million in Bitcoin, allowing zero-fee off-chain transactions between content creators and consumers.


I don't understand why Mozilla doesn't implement a native web3 wallet directly in Firefox. This would help onboard people more easily to Ethereum & EVM-compatible decentralized networks. People wouldn't have to install MetaMask or download another desktop wallet/browser app.

And then the lightning crowd would ask why did you not implement their wallet... Or say I want to do payments on the web using my creditcard,paypal. You need to have something built into the browser that allows transfer of value be agnostic to underlying ledger

Congrats Stefan & team!!!

Very curious to see how this will shape up.

Would you say this directly competes with Brendan Eich's BAT token in Brave? Yes/no? How so?

I've also been working on figuring out some mathematical models for economically evaluating digital goods (intro here https://hackernoon.com/wealth-a-new-era-of-economics-ce8acd7... ), would love to discuss various web/open-source centric economics ideas with people!

So, if I read this right, they want privacy-focused tech hipsters to vie for $20M a year to create open source fintech startups that will beat global banks and payment companies on transfer fees while simultaneously improving all social classes' access to the internet.

Here's my solution: chargebacks. You surf a site, and see a preview of some content. To get the full content, you click "buy", and the site communicates with your own preferred online payment processor (OPP) to confirm a payment. The OPP checks your balance and replies if you have the balance requested, and the site then serves up your content. At the end of the month the OPP settles up with each site for what's owed. If you want to get fancy you can implement a whole rules engine so each site can set up subscriptions for users so users can limit how much each site can charge them. You support multiple subscription models so users can pay by a monthly contract or per-click. If everyone keeps their money in the OPPs (ala Paypal) the OPPs end up making tons of money by investing the balance that hasn't been transferred out yet.

Once this is implemented, Apple, Google, Amazon and Microsoft will become the default OPP, because they still control all the platforms. Hooray open standards.

To all the people saying some variant of "there's nothing new here", I think you are missing it: Tipping has always been an effort. If it all happened behind the scenes, with no action required on your part (beyond just browsing to a site), then "tipping" can viably compete with ads.

They would do better if they spent 100M on ads to promote Firefox downloads.

Privacy is a lost cause. Sure, you'll probably get anonymity ie. your PII are gonna be scrambled, but all our activities are gonna be streamed to AI engines for commercial and entertainment purpose.

And people are generally okay with that because said engines are producing a lot of value in terms of targeted ads, better suggestions etc.

What's going on at the moment is that governments are feeling left out because they wouldn't mind running their own AI engines to shape public discourse and opinions.

But the trouble is - they don't have data. The big tech has data so I believe fight for data access is underlying reason for anti big tech momentum building up over the past few years.

Instead of having payment systems that seem to go out of their way to make it inconvenient to pay, there should be a small number of very well defined friction points (such as: “maximum daily expenditure”) with literally zero friction anywhere else.

By now for instance we should really have a security technology that allows anyone to spend $5 a day any way they want, in any denomination (e.g. literally 2 cents here and 2 cents there), without having to go through ridiculous virtual-cash systems, cards and other hoops. The money should simply be there, and it should be able to go anywhere. If it’s a relatively small amount, it doesn’t require a lot of gate-keeping, e.g. you do not need a customer-refunds infrastructure or fraud protection scheme for pennies. The money can then either reset automatically, e.g. each day, or after some explicit more-secure action, e.g. you log in to something with your face.

I know that it would personally make me spend a lot more if a lot of transactions could “just happen”. Once I got a tap-here-and-pay watch, I spent more. Once I could shop online with a fingerprint, I actually did shop online. When I had to type in credit card details though, I hardly ever wanted to. When “fees” would appear as part of seemingly-small transactions, I hardly ever wanted to continue. Some things just sound like scams (don’t get me started on subscriptions), making me more likely to find an exit.

Having worked in ad-tech for much of the past decade, I don't really like the idea of ads, but the reality is that advertising is much too effective to simply "go away." The article also conflates "advertising" with "data mining" -- the former is relatively benign; the latter not so much. The effectiveness of Facebook and Google ads are mainly due to (for the most part) information that was voluntarily handed over (likes, search history, etc.), not some evil rootkit.

And even though it sounds like a lot, 100 million dollars is a drop in the bucket. Consider that the top YouTubers make around ~$20M a year (this money comes largely via advertising). So your fund would barely cover the top 5. Also, the idea of lowering "the administrative costs of receiving payment for web content" seems kind at odds with a bureaucratic "Advisory Council initially made up of representatives from Coil, Mozilla, and Creative Commons."

I dunno', to me it just seems like posturing. No real solution offered, just some wishy-washy "feel good" proposals. At least Brave (which is my daily driver) took a hard stance against ads and blocks just about everything, while Mozilla just talks a good game[1].

[1] https://twitter.com/jonathansampson/status/11658588961766604...

It seems like a good chunk of money to help startups get started?

If a payment system can't make money or raise more funds, they're probably not getting anywhere.

Advertising is opt-out. Every idea centered around tipping, micro-transactions, cryptocurrency are opt-in. I don't think most content creators are willing to make that jump.

I know credit card processors are hostile to "adult content", but as a human being with a sexuality that's part of who I am, I'm saddened that even "literature" related to sexuality is banned.

The grant says it's seeking "makers of all stripes: webdevs, game developers, bloggers, photographers, musicians, journalists, filmmakers, writers, podcasters, software developers, and more"[1]

But Coil forbids "Pornography and other obscene materials (including literature, imagery and other media) depicting nudity or explicitly sexual acts; sites offering any sexually-related services such as prostitution, escorts, adult pay-per view, adult live chat features; sexually oriented items (e.g., adult toys); adult video stores and sexually oriented massage parlors; gentleman’s clubs, topless bars, and strip clubs; and sexually oriented dating services."[2]

[1] https://www.grantfortheweb.org/ [2] https://coil.com/terms

They could help also to build/improve the OpenCollective[1] that is a completely free and open-source[2] alternative to Patreon, aiming mostly for the open-source projects.

[1] https://opencollective.com/

[2] https://github.com/opencollective/

Perhaps this project will receive a grant then to help it improve?

Interesting that Ripple affiliates are sponsoring this.

Ripple is sponsoring this.

They gave ~$250m (in XRP) to Coil[0]. Coil is now giving $100m to people to monetize the web.

Gotta do something with that sweet, sweet ICO money.

[0] https://www.coindesk.com/ripple-is-giving-away-1-billion-xrp...

Is this right?


It lists Coil as a provider.

Here's an idea that just came to me. Feel free to shoot holes in it. It's based on my assumption that users do prefer things without ads, and might want to pay for it.

So how about a search engine where, if you pay for it, it ranks sites without ads at the top. And if you click a link to an ad-less site, that site gets some of the money you paid to the search engine.

Maybe combine it with a social network that works a similar way, though that's harder to do because you don't control what people share.

The major downside is that there would be a central place (the search engine) where the money gets distributed. It would be better to have something non-centralised. Still, I might use a search engine like that.

A major question, for me, is: how does it get started? What would motivate someone to create a website without ads, for the purposes of what you're suggesting, when the scheme is in its infancy? What would motivate a website to remove ads, when the scheme is in its infancy?

Plenty of websites already have no ads. Wikipedia, for example. Many blogs too.

But even with ads, you'll still be listed. But ad-free sites get a higher ranking, making that a more viable decision for some.

That doesn't matter unless the existing pool of ad-free sites can supply good results for a reasonable proportion of the actual web searches being performed. I doubt very much it can.

For the idea to work it'd mean that there'd b incentive to have a website on a particular topic be ad-free vs ad-based competitors. How does that incentive start in the first place?

It probably shouldn't put all ad-free sites above all ad-laden sites, but some preferential treatment would be nice. But there are already plenty of ad-free sites that have tons of useful information. I wouldn't mind seeing them ranked higher.

I can’t help but think that this is an example of jumping to the wrong solution too quickly. Is the problem that the web doesn’t contain as much high-quality content anymore as before; or that independent content providers are having a hard time making a living. I don’t enjoy the web as much as I used to, but I am pretty sure that in the early days almost nobody was making money off the content they created. So maybe the problem is discoverability rather than compensation. It would be helpful to see a little more of the analysis that Mozilla has done to define the problem.

Want to help publishers? Fund a competitor to google that either doesnt track users, or does it in an ethical way. With monopolistic google disappointing both advertisers and publishers by raising prices and keeping most of the profit, there is an opportunity in the market that wasn't there 10 years ago when most of their competitors gave up.

Non-ad monetization models are not hindered by lack of investment or ideas, they are hindered by REGULATION. The best ideas are impossible to do with today's regulatory and fee structures, and publishers in particular cannot escape that.

Would 100 million be anywhere remotely close to enough for that?

Bing, with all the money and power of MS behind it, has about 5% market share. Yahoo has about 2% market share. Baidu has about 15% but also total assets of 40 Billion US Dollars.

> The best ideas

What are those?

> Would 100 million be anywhere remotely close to enough for that?

I think money is not the biggest hurdle. The problem with most ad networks is they copy google: most of the adsense alternatives for example are simply googles's ad partners that use google's ad exchanges. I think the ad space instead needs a "stripe for ads", a company that will quickly catch up with what publishers want, create solutions and push advertisers to use them.

The alternative ideas include anonymous micropauyments.

> Non-ad monetization models are not hindered by lack of investment or ideas, they are hindered by REGULATION. The best ideas are impossible to do with today's regulatory and fee structures [...]

Can you say more about this?

having thousands of customers paying $0.1 per day is a huge accounting hassle. Trying to account e.g. for EU VAT is a maze of galactic proportions for a small startup. So you 'll need an intermediary who will get a cut of that income. And then you 'll occasionally be getting weird looks from banks because apparently that business model looks suspicious ( i've had my bank deny me services because we do virtual currency sales in game - through an intermediary).

There is also no frictionless way to do it: payments should ideally be integrated in browsers, and people should be able to pay anonymously by preloading their browser with small chunks of money, again anonymously. Maybe there is a way to make this mechanism both anonymous and legal (for tax purposes) , but i don't think anyone has cracked that yet.

Advertising circumvents all these requirements, and creates an indirect channel that transfers value from the user to the publisher. And advertising is here to stay: businesses always have and always will will keep spending a similar % of revenues for advertising. It's not advertising that's wrong with the internet, it's tracking.

I wonder if a Spotify for the Internet would work.

Pay a monthly access fee and have micro payments go to the creators based on usage.

Mozilla would be in a good position to introduce that type of service and track consumption.

> Pay a monthly access fee and have micro payments go to the creators based on usage.

That's exactly how Brave Browser works.

This would be so neat but a problem is how do you distribute money in a fair way to creators? Spotify can give X dollars to a song author for each minute a user has listened to his song. You cannot easily do that with web content.

I see the following models of monetizing content on the web:

1. Gift economy (you give it away, take likes) 2. Goods and Service economy (you charge for it, sufficiently to make a living) 3. Advertising economy (many people want it but not many willing to pay for it) 4. Tipping (not a viable economic model, unless it's made compulsory and a minimum is set in which case it's not tipping)

1 & 2 have been with us since the dawn of human civilization.

3 is a modern invention and to many it is a failed experiment.

Just observations, no answers

What makes advertising a failed experiment exactly?

It's currently the only sustaining model for 95% of digital content and has created at least one trillion dollar corporation. That's the opposite of failure.

One good thing about advertising is that it allows the poorest free access to most content. Hopefully these initiatives will keep that.

Trying to divorce publishing from advertising is like trying to divorce brewing from yeast. The only people who don't understand this are people who have never been in the publishing business.

If you think about this as a marketing campaign for Coil then it seems to me to be brilliantly enlightened. I mean that sincerely, no sarcasm.

Here's how Coil ended up with $100 million to fund this grant (https://fortune.com/2019/09/16/grant-for-the-web-mozilla-coi...):

> As for the source of the $100 million that will fund the project, Thomas said it came via a grant to Coil from Ripple, and that in some cases, the project will use XRP—the cryptocurrency that is an integral part of Ripple's operations—to settle financial transactions.

… and here's why Ripple made the original (~$265 million USD) grant to Coil (https://www.prnewswire.com/news-releases/ripples-xpring-make...) in August 2019:

> Coil, a platform dedicated to reimagining monetization on the web for creators and their fans, today announced a 1 billion XRP grant from Ripple's Xpring. The money will be deployed towards driving adoption of XRP and the Interledger Protocol (ILP) by growing Coil's monetization platform through mainstream adoption of Web Monetization, an open web standard built on Interledger that enables streaming micropayments in any currency, including XRP.

> …

> The platform enables creators to post public and exclusive content on Coil, which is automatically enabled for streaming payments. Creators can web-monetize their own websites by adding a simple tag. Those who want to support creators using Coil can join the community with a $5 monthly subscription. There are no membership fees for creators.

In May 2018, the first of several lawsuits was filed that allege XRP is a security: https://www.coindesk.com/investors-suing-ripple-cite-sec-gui.... They're seeking class-action status to represent everyone who bought XRP. Matt Levine briefly mentioned Ripple while covering the broader question of how the SEC would classify coins: https://www.bloomberg.com/opinion/articles/2018-06-15/the-se....

Obviously Ripple is inherently motivated to see XRP adopted, but also, an active micropayment/content payment ecosystem might strengthen their claim that XRP is a utility coin and that their ICO wasn't an unregistered securities offering. Some background on this question: https://www.sec.gov/news/public-statement/statement-clayton-...

Here's what that lawsuit alleges (http://static.coindesk.com/wp-content/uploads/2019/08/716bee..., as amended Aug 2019):

> Ripple claims that XRP has utility—like currency—in its use as a “bridge currency” for international payments. But, as discussed above, more than 60 percent of XRP is owned by Ripple and none of that XRP is used for anything at all, other than to be sold in the future to investors. Moreover, as for the XRP that was already sold or otherwise distributed by Defendants, the vast majority of it is not used for bridging international transactions, but for investment purpose. Accordingly, Defendants’ claim that XRP has a utilitarian purpose is nothing but a red herring attempt to avoid the application of securities laws.

Don't forget that Coil was started by the Ex-CTO of Ripple.

Ripple never had an ICO - check your facts

Ripple doesn't hold 60% they hold 50%

The purpose of the 50% they still hold is public know since the very first days. They SHOULD use it to push the adoption of the XRP Ledger. Period. People hate Ripple for selling/donating their XRPs, funding great stuff like Coil with it that obviously somehow uses XRP but that is exactly what they HAVE to do. Heck the whole company was made for that. The XRPL devs made a better Bitcoin and wanted it to be used. How evil from them. /s

The whole security thing is nonsense. Someone bought XRP high and sold low got made and sued. Pathetic. He didn't even bought them from Ripple.

I’m not and don’t claim to be an expert. That’s one reason I don’t have (and didn’t give) an opinion about the lawsuit.

Regarding the lack of an ICO, you’re correct. Others have called XRP a perpetual coin offering[1] since Ripple clearly raised money from investors, but you’re correct that’s not an ICO by name.

The rest of my comment was just direct citations, so direct your sarcasm at them.

[1]: “Ripple Hit With Class-Action Suit Over ‘Never Ending ICO’”: https://www.bloomberg.com/news/articles/2018-05-04/ripple-hi...

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact