2016-02-29

An Ethical AdBlocker - a Bitcoin killer app?

Ads suck. TV ads suck, internet ads suck, pretty much all ads suck (well, perhaps except for some Super Bowl ads and a few rare gems). Moreover, ads don't work as well as they used to, so they have to get more aggressive. We hate ads, so we create software to block ads, and the software creators then charge a toll to let companies display us ads anyway.

A lot of content creators need ads to support their work, they appeal to their viewers not to use adblocks, and some websites make half-assed attempts at getting us to pay instead of viewing ads.

If you don't use adblocker, your viewing experience on most sites is abysmal, and the loading times, especially on mobile, are abysmal. If you use adblock, you feel bad for not supporting the content creators. Perhaps instead you decide to use "ethical adblocks" and only view websites that are not ad supported?

All in all, the current ad industry is a struggle between ad creators that want you to see their ads, content creators that don't want to show you the ads but have to to earn a living, and the viewers that don't want to watch the ads but still want to support the content creators.

Perhaps there is a way to support the content creators AND not have to rely on ads? There have been many approaches to this in the past, so lets see what we can learn from them and what can be improved...

Tip-based support


One of the first ideas that come to mind when one thinks how to support a creator is tipping. The idea is not new - companies like Flattr have been around for many years.

The problem with tipping, is that quite often you have to set a tipping jar up first and hope your users will be using the same platform. Without critical mass, you don't have much. This seemed to be the reason Flattr failed - it hasn't reached a critical mass, so it fizzled out mostly.

A better idea came around with BitcoinTip and later ChangeTip - tipping solutions where you could tip anyone, even if they haven't set up an account with the company. Combining that with being able to tip anyone on various social websites like Reddit, Twitter, Twitch, etc. Now suddenly you were able to even tip famous people, such as Garry Kasparov, and know they would receive their bits.

However, tip-based support also has a downside - you can't make a predictable living with them. They are by their very nature a flash in the pan. Moreover, just like upvotes on Reddit, short, witty jokes and memes might get content creators more tips / upvotes than large submissions of substance. We need something better that encourages creation of "wholesome" content, not just pictures of cats.

Patronage


An alternative to tip-based support is patronage, made especially easy through such platforms as Patreon or SatoshiVote. This form of support is suitable for a much broader set of work - from hobbyist book reviews up to high-quality educational videos.

This models fosters consistently high-quality work. Creators get paid more the more people enjoy their work on a regular basis. One-hit wonders don't translate directly into money as it would with ads on a highly-popular video, but the extra exposure can translate into a bigger stream of money down the line.

While this model is good for continuous series, it might not be ideal for infrequent releases that get a lot of views over time (say, the song "Friday" by Rebecca Black is getting a good amount of searches every week on Friday), or frequent but very minor releases (such as most of the top contributors to Reddit).

Subscription model


Similar to the patronage model, in the subscription model every user pays a certain amount per month to view a website (usually, but not always ad-free). While this model can work pretty well if the user consumes a lot of content from the same website, it doesn't work well if someone visits a page only a few times per month. Since one can only pay for so many subscriptions per month, it tends to promote only the biggest, most established content hosts (similarly to subscription-based MMOs being replaced by free-to-play ones).

A tipping adblocker


A possible solution to most of those problems could be a combination of a few of the above technologies.

The starting point for the solution would be an ad blocker - not necessarily designed to sell your data nor to take bribes to get ads through, but focusing on being the best ad blocker out there - no compromises, user experience comes first.

From there, the software would start tracking which pages its users visit and for how long. All of the time would be divided based on the content consumed - length of the video watched, which submissions on Reddit were upvoted, etc.

Once the data is aggregated, the users would be able to donate a certain amount of money per month. Like on Flattr, the donations would be divided up between the content consumed, but without the extra hassle of having to click on special buttons and so on. Some portion of the money could go to the company maintaining the whole software product.

The content creators would receive the aggregated donations of everyone from the system. If they provide a Bitcoin address - the money can be sent automatically. For websites that support it, the money could be deposited to user's account like through ChangeTip. Other websites could embed payment details directly into the html code, which can be easily scoured like ProTip appears to be doing. If a website doesn't have a payment setup yet, the donations could be stored for the time being. Websites hosting content could divide the revenue up - YouTube could get a cut of the tip, while the video creator could get the rest.

Everything could be optimized through some Interledger-like cross-currency payment protocols if someone doesn't like Bitcoin. Users could also add weight to various sites they are tipping - perhaps prioritizing blog posts over Reddit and so on.

All in all, this setup would improve the user experience by eliminating ads, replacing the revenue stream content creators and hosts lose from the removed ads by essentially donate-per-view (which could be more than a few cents an active user is generating per month). This means a user doesn't have to commit to a subscription if they don't visit a site frequently, they reward the content creator whenever they view the media, not "once-per-video" like Patreon would, and adding an extra tip every now and then would be seamlessly integrated into software (since it already tips everyone anyway, might as well make it easy to tip a bit extra).

Conclusions


Nobody wants to watch ads, thus adblockers are popular. If the adblockers would incorporate some universal tipping solution to replace the revenue stream from ads to support the content creators, it could be a win for both the creators and the users (although not so much for companies that need to advertise). Bitcoin could allow for anyone to receive tips, but different payment solutions could be introduced as well.

2016-02-16

Breaking dollar's fungibility

In the modern world we rarely make a distinction between the money in our banks, the currency in our pocket, or our balance in a digital wallet like PayPal - a dollar is a dollar, pretty much fungible. However, that's not really the case - money deposited in a bank means the bank owes that money to you (it's not "your cash"), digital wallets also own your money and can easily freeze your balance. Transferring money from one bank to another is always done at par, even if that bank in question might be Lehman Brothers about to go down back in 2008. Perhaps it is time we break the dollar's fungibility and start putting a price tag on the credibility of banks?

Private notes


Over 150 years ago in the "free banking era", any bank could issue its own banknotes. You would see a number of different notes in circulation - you could have $2 from The Bank of Chattanooga, The County of Polk, or The Lawrenceburg Bank of Tennesee. Same in Canada. While it created a lot of hassle for anyone wanting to use the currency, especially if they would travel beyond where those notes would be redeemable, it also allowed for a market to form and put a real value on the currency based on how credible the issuing bank was - good notes would be valued at par, bad ones - at a discount.

While banking nowadays is certainly simpler and safer with uniform currency issued by one entity per country, FDIC deposit insurance to prevent people losing money in case a bank goes bust, etc. However, this means we also lost the ability to evaluate bank's trustworthiness, usability, etc. and arbitrage it.

Bank arbitrage


In the Bitcoin world, it is fairly straightforward to judge the health of an exchange by looking at its exchange rate. Back when MtGox was going bust, its rates deviated from its competitors by 15+% even early on. When withdrawals out of the exchange became impossible, there was even a secondary market that traded MtGox BTC for real BTC by using MtGox's inter-account transfer capabilities. You can track arbitrage metrics today.

How would this apply to banks? Well, you could start with the currency issued by the government as the base - one dollar here would be redeemable to one dollar in banknotes (this is what MintChip aimed to do for example). Banks would use that as their reserves for fractional-reserve banking and issue their own debt-based currency. All of it could be tracked on a shared "bankchain" to allow market for various bank debt to form. The price difference of the debt could stem from various factors - how stable the bank is (FDIC insurance is all well and good, but nobody wants to go through the stress of having your saving locked up for who knows how long), how cheap and easy it is to transact to and from a bank (say, USD-demoniated bank account in Europe might be valued less due to the extra cost of transferring money overseas), how accessible is the bank (branch opening hours and how common they are), as well as how their customers share the banks values (banking for millennials, sharia-compliant finance, etc.).

Setting up such simple metric for each bank would allow anyone to easily compare various banks and put that metric on everyone's mind. If one day your money would go down to 95 cents on the dollar, perhaps you would ask your bank "what is going on?" and find out that HSBC enabled Mexican drug cartels to launder money. Maybe it would enable some people to demand all of their deposits to be covered 100% by the government-issued currency, rather than allow for fractional reserve banking? Or perhaps it would allow some people to move their money to their local credit union to support the grassroot company and earn 1-2% on the conversion rate.

Conclusions


Banking in the modern world is pretty homogeneous - currency dictated by the government, fungible money no matter where you go. Perhaps it might be useful to bring the market back into the equation and allow us to see see the bank's worth by checking the value of their dollars?


Related links:


2016-02-08

The Dark Wiki

This blog post is inspired by some "what if" videos made by Tom Scott, in which he explores a theoretical futures when Google forgot to check passwords, the dystopian view of the singularity ruined by lawyers or what happens when privacy dies. Here is my take on what might be built in the future on top of the existing blockchain technology.

Any names or examples used in this story are meant for purely illustrative purposes only and are not meant to condone or condemn any actions, companies, governments, technologies or the like.

----

Knowledge is power, and a lot of money can buy you a lot of information...

The years is 2025, there are almost 20 million bitcoins in circulation and they are valued at $5'000 per coin. While not as fast a growth as many would've predicted, it still puts Bitcoin as the world's 8th biggest currency in circulation, ahead of Canada's dollar, but behind the United Kingdom's pound. For all intents and purposes, Bitcoin and crypto have succeeded - they are used by a lot of people in all walks of life.

The biggest story of the year's first quarter turned out to be the busting of a US-based criminal organization creating superbills - high quality replicas of the "counterfeit-proof" polymer banknotes. What stood out about those bills is not that they were good, but that they were perfect replicas of the banknotes, down to the microprint, security ribbons and the colorshifting ink.

While not an unheard of story (North Korea is said to have done something similar in the past), over the following days of media attention a surprising finding has caught everyone's attention - the criminals got all of the information they needed to make the superbills from "The Dark Wiki" - a Wikipedia-like website residing on the Dark Web, created with the sole purpose of "cataloguing the world's forbidden knowledge".

The Dark Wiki contained detailed articles on many subjects that would land anyone in jail - recipes for narcotics, 3D printer files for military-grade weapons, interrogation manuals from a number of international agencies, copies of standard keys used in "back doors", the complete smallpox genome, or "how-to's" on accessing military GPS signals or the aforementioned printing process of superbills, from 3D prints of various master hubs, etc.

The website was not only dealing with standard articles, but also featured a prominent section on "kickstarting" / requesting information, with the bounties anywhere between a few thousand dollars for creating some malicious scripts to break into some computers, through a few hundred thousand dollars for pharmaceutical recipes of some major drugs, up to a few million dollars for schematics of nuclear weapons.

While the website seemed to lack a single owner, there were a number of prominent "experts" on the website serving as third party escrows on a number of bounties to verify if the information delivered was accurate and precise. They also curated one of the more bizarre part of the website - a scientific journal on "the dark sciences".

While a number of traditional scientific papers have been written on criminals and their illegal activities (such as Steven Levitt and Sudhir Venkatesh studying the economics of crack dealing), the Dark Wiki's journal was focused more on scientific research that was itself illegal or borderline illegal - designer drugs, human cloning, synthetic organisms, or experimenting on humans and infants. While so far only the first category had any traction, going so far as having some research grants, the other categories were open for submissions.

With a little help from the Streisand effect, hundreds of millions of people have heard of the website and it became the most searched for topic of the moment. Among the wave of new users that came across the website for the first time then, was a disgruntled 30-something NSA software engineer. Stuck in a dead-end job working on some machination that would further make the agency's reputation worse if it ever was revealed.

What he saw in that website struck a chord with him. When he was a teenager, Wikileaks broke the news. In his early 20s, Snowden's revelations shook the world. This year, he would make both history, and a lot of money for himself.

A month later, the Dark Wiki broke into the news again. A high six-figure bounty was awarded for delivering the information required to access NSA backdoors embedded in billions of devices world-wide. That day would later be known as "the day the Internet broke". The massive-scale hacking that occurred with that information was an order of magnitude higher than the 2014 Heartbleed bug, and it brought down a number of key servers around the world for a few hours before some of the vulnerabilities could be patched and the Internet reconnected.

The following days the governments were dealing with a few major issues. First, any US-based manufacturer was distrusted overnight, with many world governments revving up production on their non-backdoored hardware. But due to the nature of how hardware is manufactured, it would take years for the industry to adjust.

Second major issue was figuring out who leaked the information. The bounty was paid in bitcoin, but since the 2019 halvening hard fork introduced confidential transactions to the network, it became impossible to figure out which address received the funds. An internal investigation was under way at various government agencies to see if some leads could be found...

Lastly, the Dark Wiki had to be shut down - it attracted too much unwanted attention and could cause global security concerns if it was used by large criminal organizations or corrupt governments. With some busy work and exploiting some weaknesses of the Tor network, the website was finally traced to a server sitting in some country with abundant legal loopholes and lax prosecution laws for hosting illegal content. However, the biggest surprise came when it turned out there were only two notable pieces of software on the server - a small wrapper serving the website's frontend and an Ethereum client...

As it turns out, the website was running as a smart contract on the Ethereum platform. All the required information and logic was stored in the distributed ledger, with payments being handled through a sidechain connection onto the Bitcoin network. Editing rights were only given to users that created a high enough proof-of-burn pledges (by donating the coins to the contract itself) and the little governance there was was mostly handled through anonymous users building their reputation by contributing to the website and being recognized by the community. Anonymous donations from people benefiting from the wiki as well as the pledges and other fees allowed for the contract to become a self-sustaining DAO.

After the website was shut down, multiple other mirrors cropped up along with the source code required to access the data locally. Attempts to shut the network down only strengthen it due to the antifragile nature of crypto.

On the 10th anniversary of its genesis block creation Ethereum was no longer seen as a quirky distributed state machine, but as an avatar of unstoppable quest for knowledge in all of its forms paired with the cold machinations of cryptographically untouchable capitalism.

If the Dark Wiki has taught us anything in its following years of operations is that security through obscurity is a joke and that we can't rely on any secrets to keep us safe. In the end, backdoors had to be closed, strong cryptography became the default and the attitude towards cryptocurrencies has changed. Just like you cannot kill an idea, neither can you stop a decentralized network or currency.

----

What I have described above could be implemented today, at least technology-wise. It would probably still take some time before the technology becomes popular enough for people to start using it as described, but we have some kernels of that already in the form of the assassination market or the now-defunct Silk Road. In due time perhaps we would see some dark web information market develop and turn into some wiki for people to use. Pair cryptocurrencies with the sort of money drug lords or a small government can amass and you might have to be keeping a closer eye on disgruntled or low-paid employees with access to secret information...

I used Ethereum as an example of a platform to host the Dark Wiki, but the same project could be hosten on any platform that supports smart contracts. Please don't read it as condemnation of Ethereum either - I see it as one of the more innovative crypto projects and see it being useful for a number of good projects in the near and far future.