Web Monetization - Using Coil and Uphold to Combat the Network Effect
Previously titled: "The Internet, Subscriptions, Big Winners, the Networking Effect, Makers & Creators, and Society Today"
Posted on June 19, 2020
This post has two parts.
Part I is about the culture around payments, donations, and transactions in general online - and what could be a revolution and explosion in creativity. It was originally what I thought would be the complete post and drafted in October 2018, titled The Internet, Subscriptions, Big Winners, the Networking Effect, Makers & Creators, and Society Today, but because I wrote myself into a confusing economics and market based conundrum, I didn't see a way I could publish it. Then, in June 2020, some happenings on the internet had solved the conundrum for me!
Part II is about how you can use Uphold and Coil as a method to reward creators - as a way to battle my complaints and issues I raise in Part I. (Ironically, while Part II was not originally in the original draft, it is actually the very solution as to what I thought was hopeless in Part I, and was finally the motivation I needed to officially publish this post in June 2020.)
Whew. Still with me? Let's go!
Part I: Ads, Companies, Creators, Makers, the Network Effect, and Redistribution
I've been mulling over this thread of thought for months now. After enough rumination, I've finally been able to pull it all together into this fairly large post. I've collected a variety of fairly recent sources that convey the fact that these issues have been trending for a while and are coming to a point in the digital economy[^1].
I'll start off with what is going on with creators or creative entities - whether companies or individual Makers (all credit to Product Hunt for that term 😉), and then shifting to how everything appears to everyone on the other side so to speak, i.e. consumers in 2018, 2019, and 2020. I finish with some suggested solutions to the problem, and things that I am continuing to consider, even after distilling my thoughts into this post.
This is a longer, non-technical post of mine, but I think these are very important issues that need to be discussed more often. So grab a :coffee: and put your thinking cap on!
Advertisements and the Network Effect Perspective of Companies, Creators and Makers
This paper about the role of luck was one of the first things that got me thinking along the lines of these issues. While the paper is aimed at single individuals, if a company or organization is considered as an individual in the larger marketplace, many of the same findings and principles can apply.
The study found that counterintuitively that people of only average intelligence achieve the most success, on average. Unfortunately, most intelligent (or even most hard working!) were not those who were most successful (granted, there was a slight skew of more intelligent people able to achieve better on average, but overwhelmingly they were not the most successful.)
How is this possible? As succinct as I can make it, it comes down to distribution overlayed with exponential returns. The spectrum of hardworking and / or intelligence in a given population follows a normal distribution (i.e. picture the classic bell curve). Returns on success (or how well you can capitalize on a given success, as the study sometimes put it) return in an exponential manner. This exponential overlay is what leads to those of average intelligence or work ethic resulting in the most. If you give the same chances at success to all members of a population equally, then it is logical that the largest section of the population (i.e. the most average) will turn out to be on top. In other words, it's luck.
Since the researchers had of course total control over the experiment, as the second half of the paper, they examined the resulting overall success of individuals in the system if a refunding process was implemented. They tested a number of patterns, with just a few patterns listed here:
- an 'elitist' policy, in which 50% of the best performs are rewarded for each success in addition to the success itself, which only increased the top success rates of the top 50% slightly
- a 'mixed' policy, in which part of the policy followed the 'elitist' policy, but 25% of funds were redistributed to all other individuals regardless of success or failure
- an 'egalitarian' policy, where all participants receive equal reward for every opportunity regardless of success or failure, in which the average success level of all individuals increased by 17.78%.
The authors call the ability for a refunding or distribution policy to raise the success level of an individual the 'efficiency'. As could be expected, the most 'efficient' policy was that of the egalitarian policy. It seems to me this is in a way a direct mathematical proof of the concept "a rising tide lifts all boats". That is, if the average population is the tide, sharing the wealth to all equally raises everyone up in the long run.
This redistribution is exactly what is missing on the internet and the core issue I am going to get at in this post.
But wait you say, "the internet is already by definition a giant sharing and redistribution network!" This is true, but let's be honest, it's the sharing and redistribution of information, not money or wealth, as in the paper. I believe these two things, information and money, are imperative to reorganize in order to improve the environment surrounding online creators and makers.
The Subscription Based Web
To understand why redistribution hasn't been monetary based and only information based, we have to look at the companies that define huge percentages of the internet right now.
Let's look at a concrete example by way of monthly subscription style services I've been thinking of subscribing to, but haven't because I'm too much of a cheapo:
- $4 / mo. New Yorker Subscription ("just" $1 a week as their add states!!! 😑)
- $5 / mo. Medium membership (yeah, I guess the free version is not enough for me - I read a lot of people's stuff)
- $6.67 / mo. The Economist (subscription itself is 12 weeks = 3 months, for $20)
- $5 / mo. Sam Harris's Waking Up Podcast
- $5.99 / mo. Reddit Premium
So... that adds up to nearly $21 a month, or $252 a year! $252 gone, just for five subscriptions, and this is a conservative list - these are just the subscriptions that I'm toying with / have to deal with working around.
Since I'm currently debating if I myself should subscribe to these services - I realized I, myself am a self-fulfilling producer of the problem of this monetization issue!
I don't argue that $1 a month or even $5 a month is too much for a single service. (Though it starts as soon as you start consider 5 or so or more a month!) What is more troublesome is that these revenues go to big companies that can then recycle their revenue into more advertising and better development. This is the network effect at work. Then these companies grow, and, if they go public, in order to grow their numbers to make Wall Street happy, they have to resort to one thing to keep growth metrics alive:
They use ads!
The Ad-Death of the Modern Social Network Company
See: The heat death of the universeA joke I like telling is picturing the end result of Facebook, Instagram, Twitter, and other socials far into the future. They will, in my opinion appear to simply be a constant stream of - not of what is going on with your family and friends' life - but ads!
This has to be the case. They either continually grow revenue (which, as far as I know, the overwhelming majority of revenue from these social networks is ads - so that means more ads). The only alternative is to level off or even decrease the amount ads. But these companies can't do the latter option - that would cause their revenue to stagnate or shrink. Egads! We've seen what happens as soon as Wall Street even gets wind of a decrease in anything growth related!
The Consumer's Perspective
The current outlook of those who are supposed to support the digital marketplace - the consumer at large - is just as grim. Again, I'll use Facebook since it is such a prime example. I haven't used Facebook for three years now[^2], but I was looking over the shoulder of my buddy when we were hanging out this summer[^3] while he checking out his Facebook feed, and immediately exclaimed, astounded: "What the hell happened?! It's all ads!"
My buddy, sounding jaded and already much used to this ad deluge, simply replied "Yep, Facebook is just all ads now, dude.".
My brother, graduating with his bachelor's degree this year (so just inside the start generation Z), says that he and his friends use facebook only mostly for events now, also because of this über ad attack reason.
Another Maker, Justin Jackson brings up another point in this podcast episode is that with all these ads, the user is overwhelmed, and in Justin's words 'exhausted' from the constant bombardment of the next big thing. It's training many consumers to simply ignore all those ads they constantly have to scroll through, regardless of if the product itself could perhaps deserve a look or two.
It's true: while the consumer these days is empowered with making their own decision, this requires that products be very useful, very unique, and with stories behind them, to really punch through the average person's ad ignorance skills. Printing some text on a t-shirt isn't going to cut it in this next generation of the economy.
Hope for Non-commercial Social Networks
What encourages me is what I've seen through Hacktoberfest, in that groups of passionate developers (the DEV community is a great example) are already toying with ways of creating social networks that are complete open source - funded by donations, sponsorships, or the developers themselves - thus ad free (or at the very most, relevant and non-spammy ads)! As always in tech, internet tools such as hosting and storage are getting more and more affordable, so much so that a group of individuals (or in some cases even a single individual!) can afford to run an app that is used around the world.
In these types of Social Web Apps 3.0 as I hope to call them, the ads, if any, on those sites would be a means to end: they keep the service running. In this case, the model is that members of the network come first, whereas (nowadays on Twitter and Facebook), it seems like the opposite is true: the user is bombarded with ads that, while it may indeed be targeted to the their interests, is nonetheless approaching spam levels. Advertisers first, users second.
Another hope is that the new generation Z is more charitably and socially aware than any generation that has come before them. They are looking now, not just for companies and individuals that donate profits for example, but actively participate in charity work and charitable causes. I suppose from a Maker perspective, these are people who are actively contributing to online communities, podcasts, and blog posts, sharing their ideas and putting in the hard work to stand out in the digital age. Or, like DEV.to - not a bombardment of scrolling through ads - just solid and software relevant ones that keep the company running!
The Solution(s)
A "simple" "solution" might be to just post your PayPal or favorite cryptocurrency wallet address[^4], and hope that some people donate. I've checked multiple blogs that posted cryptocurrency wallet addresses and I still have yet to find one with a single transaction). The problem with this type of solution is that it requires an extra action on the consumer's part. Things like buymeacoffee.com make this even easier as a consumer (and you'll notice I also have that link here), but at the end of the day still require that initiative on the consumers part.
The best that I could think up was some sort of passive system that pays just by people being on your site, and indeed, there were groups trying to build such microtransaction systems (yes, even before crypto currency) ever since the birth of the internet age. I think I saw something on Hacker News somewhere a while back about it, but the fact that I can't dig up a reference or find a bookmark that perhaps I never made reflects how niche these attempts likely were.
The Dawn of the 'Redistribution Company'?
There's another neat way to solve this. Let's make an example:
Chris sells widgets. Each widget Chris sells would normally provide him with $10 in profit. But what if Chris redistributes the profits from every widget sold to anyone who has every bought the widget - including Chris himself! Of course the profits I realize will become marginally smaller the more people that buy my 'widget'... so, is this the very definition of a pyramid scheme? I don't think so, because I'm distrubiting everything equally. The only problem is still the issue of early adopters - they will benefit financially the most, while some of the people to purchase the product last. To some this may sound eerily 'communist'.
But isn't the alternative worse? A group of companies reaping a massive amount of the internet's wealth (Okay, to be fair, according to Bruce Upbin at Forbes and three systems theorists from the Swiss Federal Institute of Technology, 40% of the wealth is controlled by 147 companies, and 80% if controlled by 737. Meanwhile, 10 companies control 80% of the media.
Anyway, lets do the math on such a scenario. I think it's a bit weird to immediately get the refund back from your purchase, so lets consider the redistribution at the end of certain time frame - say a year or six months.
Still, this has the same structure as a corporate structure, also showing similar effects like the cryptocurrency phase, in that early adopters benefitted overwhelming the most. It all goes back to the study on luck. That lucky individuals are simply that - lucky, while only more lucky individuals reflect that wealth and success scale in the geometrically.
Still, I think one of these methods is in the right direction of creative and Maker wealth redistribution in my dream of a 'new economy'.
You're probably thinking at this point, "this guy is crazy; I'm not going to read his blog anymore". Good ma'am or sir, I laugh at you! :monocle_face: In all seriousness, to me, the prospect of having a huge economy where experts of all shapes and sizes can thrive based off donations and commission work is the ideal. Indeed, the Maker movement, Digital Nomads, the exploding size of the freelance workforce are all sounding like the markets are already stepping in that direction anyway.
Additionally, as I've frequently stated, being a software developer who is interested and always willing to learn in 2018[^5], I will never be short of having a 'job that pays the bills', although the vision in this post is to create an economy where that wont be needed anyway. 😉
My Pledge (EDIT: not really though, this section is simply preserved for history) [^6]
Posting this giant post with all talk and no action would be pretty lame. Therefore, as of today ~~90% of all donations to this blog will be recycled into the maker and creative economy~~ that I deem valid to support. These include:
- Dan Carlin's Hardcore History (Which he made free and by volunteer donation only - excellent!)
- The GatsbyJS Team - fantastic work they are doing - also and example of the full tool which is totally free and donations are optional
- The
create-react-app
Team - another invaluable tool for developers like me who use React.
~~I'm building out a page to track these recycled cash flows. I'm paying it forward in hopes to speed up my dream of a Maker, Consultant, and Creative economy a reality.~~
Future Considerations and More Work
This is an issue that I repeatedly think about, and will likely blog about again.
As I keep trying to build out an arsenal of projects as a Maker (all credit to (https://producthunt.com)[Product Hunt] for that term), I think I will be opting for the one time charge - or in the worst case, a yearly subscription.
A common complaint by software engineers is that they don't work on projects that are challenging and ground breaking, or for companies that by nature aren't pushing the cutting edge. This issue was discussed in depth in a highly upvoted 2018 Hacker News thread.
That the very nature of the network effect is a root cause of this issue.
If we want smart people to have the freedom and security to work freely on large scale 'moonshot' projects (regardless of discipline - this applies to the arts, sciences, engineering, software - you name it) we have to be more open to supporting creators, and giving them more room and security to try for those lofty goals. I'll admit myself that I typically spend 70-80% of my time working on customer's projects - because they are the projects that pay! But these projects aren't game-changing or ground-breaking in any sense. They simply maintain the status quo - "keep my e-shop running" or "streamline my invoice business process" or "paginate my blog".
I picture a world of networked freelancers all striving, while also volunteering time to contribute to open source versions of the things they are building - that the community at large can build off of. Perhaps I'm missing something that isn't feasible, but to me, such an ecosystem would be the solution of all the problems raised in this post.
Part II: Use Uphold and Coil to Support Creators and Makers via Automated Microtransactions
I couldn't believe what I was reading when DEV.to posted their "DEV is now Web Monetized" post. I was further flabbergasted when I learned YouTube was using the same technology, and even further flabbergasted to learn it is in the W3C standard and will even be finalized in mid 2020. It seems like my initial ruminations back in 2018 had been the similar thought process at these other organizations!
Promoting Simply Higher Quantity, Lower Quality Content?
But wait, you ask, "won't this just incentivize simply a larger amount of 'stuff' - 'stuff' that is low quality?". I don't think so. While that notion may be true in the short term, typically blogs gain traction based on the content itself, not the quantity. Even before this monetization tool, web traffic was always determined by. Indeed, your SEO score may not correlate initially - you may be hitting lots of key words and getting hits with so called SEO hacks, but they length of time real humans spend on your site is determined by the quality of content, not the quantity. Nowadays if after a few months of getting high hits but no actual long visiting times, google will actually lower your ranking as it's a warning signal of exactly that: low quality content. If anything, the more that Web Monetization is adopted, creators will be able to receive an even stronger signal (a monetary one!), or at least an additional signal to if their work is being well received or ignored.
My Call To Action and Pledge
These creators (myself included) won't receive donations. Its extremely simple to get started:
- Subscribe to Coil for $5/month
- Install the Coil browser extension for the browser of your choice
As part of this challenge, I've already pledged myself to subscribe to Coil for at least a year: until June 2021. In 2021, I'll review and reflect on the ecosystem, and go from there. You want proof? Here's my Payment Pointer:
$ilp.uphold.com/BJGXHRXaMM8Z
Want to check that it's valid? Better send me some money to be sure! 😂 😉 You'll see I also added some bling to the footer of the blog, simply, um "borrowing" the nice animated svg from the Chrome Coil extension. 😊
Cheers! 🍺
Chris
Footnotes
[^1]: In an amazingly ironic turn of events, I got my first donation from an ABAP post on this blog *while* I was drafting this post! (Back in October 2018) However, while it gives me more hope personally, I think the challenges raised in this post are still extremely relevant and in no way comprehensively solved!
[^2]: Five years now, as of 2020.
[^3]: Summer of 2018 now, that is 😉
[^4]: Recall in 2018 when I wrote this, cryptocurrency was still a very hot topic.
[^5]: Still true in 2020. 😊
[^6]: Update February 2020: You'll notice here there is some strikethrough. I let my ideals get to me! I realized after revising this post in 2020, that I missed a key element of free markets. I thought to myself, if some amazingly kind person donates to my blog, who am I to automatically redistribute their donation? If someone's donating to a blog, its because they want to support the blogger, not to have them just redistribute to further makers and creators. I realized, if they wanted to donate to someone else, they would have done so themselves (and are free to do so)! It's not my call to make. In the end, it appears that my first presented solution - the automated microtransactions, seems to be the way to go (indeed, easy to say in hindsight as it begins to be adopted in realtime across the web in 2020).