A reading list for successful people

Elon Musk reads and talks. So, one learns. Many brilliant individuals & amazing literature.

This article recommends 3 Elon Musk novels. All of them helped me succeed. Hope they'll help you.

Douglas Adams's The Hitchhiker's Guide to the Galaxy

Page Count: 193
Rating on Goodreads: 4.23

Arthur Dent is pulled off Earth by a buddy seconds before it's razed for a cosmic motorway. The trio hitchhikes through space and gets into problems.

I initially read Hitchhiker's as a child. To evade my mum, I'd read with a flashlight under the covers. She'd scold at me for not sleeping on school nights when she found out. Oops.

The Hitchhiker's Guide to the Galaxy is lighthearted science fiction.

My favorite book quotes are:

• “Space is big. You won’t believe how vastly, hugely, mind-bogglingly big it is. I mean, you may think it’s a long way down the road to the chemist’s, but that’s just peanuts to space.”

• “Far out in the uncharted backwaters of the unfashionable end of the western spiral arm of the Galaxy lies a small unregarded yellow sun. Orbiting this at a distance of roughly ninety-two million miles is an utterly insignificant little blue-green planet whose ape-descended life forms are so amazingly primitive that they still think digital watches are a pretty neat idea.”

• “On planet Earth, man had always assumed that he was more intelligent than dolphins because he had achieved so much — the wheel, New York, wars, and so on — whilst all the dolphins had ever done was muck about in the water having a good time. But conversely, the dolphins had always believed that they were far more intelligent than man — for precisely the same reasons.”

the Sun Tzu book The Art Of War

Page Count: 273
Rating on Goodreads: 3.97

It's a classic. You may apply The Art of War's ideas to (nearly) every facet of life. Ex:

• Pick your fights.

• Keep in mind that timing is crucial.

• Create a backup plan in case something goes wrong.

• Obstacles provide us a chance to adapt and change.

This book was my first. Since then, I'm a more strategic entrepreneur. Excellent book. And read it ASAP!

My favorite book quotes are:

• “Victorious warriors win first and then go to war, while defeated warriors go to war first and then seek to win.”

• “Engage people with what they expect; it is what they are able to discern and confirms their projections. It settles them into predictable patterns of response, occupying their minds while you wait for the extraordinary moment — that which they cannot anticipate.”

• “If you know the enemy and know yourself, you need not fear the result of a hundred battles. If you know yourself but not the enemy, for every victory gained, you will also suffer a defeat. If you know neither the enemy nor yourself, you will succumb in every battle.”

Peter Thiel's book Zero to One

Page Count: 195
Rating on Goodreads: 4.18

Peter argues the best money-making strategies are typically unproven. Entrepreneurship should never have a defined path to success. Whoever says differently is lying.

Zero to One explores technology and society. Peter is a philosophy major and law school graduate, which informs the work.

Peters' ideas, depth, and intellect stood out in Zero to One. It's a top business book.

My favorite book quotes are:

• “The most valuable businesses of coming decades will be built by entrepreneurs who seek to empower people rather than try to make them obsolete.”

• “The next Bill Gates will not build an operating system. The next Larry Page or Sergey Brin won’t make a search engine. And the next Mark Zuckerberg won’t create a social network. If you are copying these guys, you aren’t learning from them.”

• “If your goal is to never make a mistake in your life, you shouldn’t look for secrets. The prospect of being lonely but right — dedicating your life to something that no one else believes in — is already hard. The prospect of being lonely and wrong can be unbearable.”

https://tomconnor.me/wp-content/uploads/2021/08/10x-Engineer-Mental-Models.pdf

I keep returning to the same mental models and tricks after writing and reading about a wide range of topics.

Top 12 mental models

12.

Survival bias - We perceive the surviving population as remarkable, yet they may have gotten there through sheer grit.

Survivorship bias affects us in many situations. Our retirement fund; the unicorn business; the winning team. We often study and imitate the last one standing. This can lead to genuine insights and performance improvements, but it can also lead us astray because the leader may just be lucky.

11.

The Helsinki Bus Theory - How to persevere Buss up!

Always display new work, and always be compared to others. Why? Easy. Keep riding. Stay on the fucking bus.

10.

Until it sticks… Turning up every day… — Artists teach engineers plenty. Quality work over a career comes from showing up every day and starting.

9.

WRAP decision making process (Heath Brothers)

Decision-making WRAP Model:

W — Widen your Options

R — Reality test your assumptions

A — Attain Distance

P — Prepare to be wrong or Right

8.

Systems for knowledge worker excellence - Todd Henry and Cal Newport write about techniques knowledge workers can employ to build a creative rhythm and do better work.

Todd Henry's FRESH framework:

1. Focus: Keep the start in mind as you wrap up.

2. Relationships: close a loop that's open.

3. Pruning is an energy.

4. Set aside time to be inspired by stimuli.

5. Hours: Spend time thinking.

7.

Black Box Thinking…..

BBT is learning from mistakes. Science has transformed the world because it constantly updates its theories in light of failures. Complexity guarantees failure. Do we learn or self-justify?

6.

The OODA Loop - Competitive advantage

O: Observe: collect the data. Figure out exactly where you are, what’s happening.

O: Orient: analyze/synthesize the data to form an accurate picture.

D: Decide: select an action from possible options

A: Action: execute the action, and return to step (1)

Boyd's approach indicates that speed and agility are about information processing, not physical reactions. They form feedback loops. More OODA loops improve speed.

5.

Know your Domain 

Leaders who try to impose order in a complex situation fail; those who set the stage, step back, and allow patterns to develop win.

https://vimeo.com/640941172?embedded=true&source=vimeo_logo&owner=11999906

4.

The Three Critical Gaps

• Information Gap - The discrepancy between what we know and what we would like to know

• Gap in Alignment - What individuals actually do as opposed to what we wish them to do

• Effects Gap - the discrepancy between our expectations and the results of our actions

3.

Theory of Constraints — The Goal  - To maximize system production, maximize bottleneck throughput.

• Goldratt creates a five-step procedure:

1. Determine the restriction

2. Improve the restriction.

3. Everything else should be based on the limitation.

4. Increase the restriction

5. Go back to step 1 Avoid letting inertia become a limitation.

Any non-constraint improvement is an illusion.

2.

Serendipity and the Adjacent Possible - Why do several amazing ideas emerge at once? How can you foster serendipity in your work?

You need specialized abilities to reach to the edge of possibilities, where you can pursue exciting tasks that will change the world. Few people do it since it takes a lot of hard work. You'll stand out if you do.

Most people simply lack the comfort with discomfort required to tackle really hard things. At some point, in other words, there’s no way getting around the necessity to clear your calendar, shut down your phone, and spend several hard days trying to make sense of the damn proof.

1.

Boundaries of failure - Rasmussen's accident model.

Rasmussen modeled this. It has economic, workload, and performance boundaries.

The economic boundary is a company's profit zone. If the lights are on, you're within the economic boundaries, but there's pressure to cut costs and do more.

Performance limit reflects system capacity. Taking shortcuts is a human desire to minimize work. This is often necessary to survive because there's always more labor.

Both push operating points toward acceptable performance. Personal or process safety, or equipment performance.

If you exceed acceptable performance, you'll push back, typically forcefully.

I read 53 novels last year and hope to do so again.

Books are best if you love learning. You get a range of perspectives, unlike podcasts and YouTube channels where you get the same ones.

Book quality varies. I've read useless books. Most books teach me something.

These 9 novels have changed my outlook in recent years. They've made me rethink what I believed or introduced me to a fresh perspective that changed my worldview.

You can order these books yourself. Or, read my summaries to learn what I've synthesized.

Enjoy!

Fooled By Randomness

Nassim Taleb worked as a Wall Street analyst. He used options trading to bet on unlikely events like stock market crashes.

Using financial models, investors predict stock prices. The models assume constant, predictable company growth.

These models base their assumptions on historical data, so they assume the future will be like the past.

Fooled By Randomness argues that the future won't be like the past. We often see impossible market crashes like 2008's housing market collapse. The world changes too quickly to use historical data: by the time we understand how it works, it's changed.

Most people don't live to see history unfold. We think our childhood world will last forever. That goes double for stable societies like the U.S., which hasn't seen major turbulence in anyone's lifetime.

Fooled By Randomness taught me to expect the unexpected. The world is deceptive and rarely works as we expect. You can't always trust your past successes or what you've learned.

Antifragile

More Taleb. Some things, like the restaurant industry and the human body, improve under conditions of volatility and turbulence.

We didn't have a word for this counterintuitive concept until Taleb wrote Antifragile. The human body (which responds to some stressors, like exercise, by getting stronger) and the restaurant industry both benefit long-term from disorder (when economic turbulence happens, bad restaurants go out of business, improving the industry as a whole).

Many human systems are designed to minimize short-term variance because humans don't understand it. By eliminating short-term variation, we increase the likelihood of a major disaster.

Once, we put out every forest fire we found. Then, dead wood piled up in forests, causing catastrophic fires.

We don't like price changes, so politicians prop up markets with stimulus packages and printing money. This leads to a bigger crash later. Two years ago, we printed a ton of money for stimulus checks, and now we have double-digit inflation.

Antifragile taught me how important Plan B is. A system with one or two major weaknesses will fail. Make large systems redundant, foolproof, and change-responsive.

Reality is broken

We dread work. Work is tedious. Right?

Wrong. Work gives many people purpose. People are happiest when working. (That's why some are workaholics.)

Factory work saps your soul, office work is boring, and working for a large company you don't believe in and that operates unethically isn't satisfying.

Jane McGonigal says in Reality Is Broken that meaningful work makes us happy. People love games because they simulate good work. McGonigal says work should be more fun.

Some think they'd be happy on a private island sipping cocktails all day. That's not true. Without anything to do, most people would be bored. Unemployed people are miserable. Many retirees die within 2 years, much more than expected.

Instead of complaining, find meaningful work. If you don't like your job, it's because you're in the wrong environment. Find the right setting.

The Lean Startup

Before the airplane was invented, Harvard scientists researched flying machines. Who knew two North Carolina weirdos would beat them?

The Wright Brothers' plane design was key. Harvard researchers were mostly theoretical, designing an airplane on paper and trying to make it fly in theory. They'd build it, test it, and it wouldn't fly.

The Wright Brothers were different. They'd build a cheap plane, test it, and it'd crash. Then they'd learn from their mistakes, build another plane, and it'd crash.

They repeated this until they fixed all the problems and one of their planes stayed aloft.

Mistakes are considered bad. On the African savannah, one mistake meant death. Even today, if you make a costly mistake at work, you'll be fired as a scapegoat. Most people avoid failing.

In reality, making mistakes is the best way to learn.

Eric Reis offers an unintuitive recipe in The Lean Startup: come up with a hypothesis, test it, and fail. Then, try again with a new hypothesis. Keep trying, learning from each failure.

This is a great startup strategy. Startups are new businesses. Startups face uncertainty. Run lots of low-cost experiments to fail, learn, and succeed.

Don't fear failing. Low-cost failure is good because you learn more from it than you lose. As long as your worst-case scenario is acceptable, risk-taking is good.

The Sovereign Individual

Today, nation-states rule the world. The UN recognizes 195 countries, and they claim almost all land outside of Antarctica.

We agree. For the past 2,000 years, much of the world's territory was ungoverned.

Why today? Because technology has created incentives for nation-states for most of the past 500 years. The logic of violence favors nation-states, according to James Dale Davidson, author of the Sovereign Individual. Governments have a lot to gain by conquering as much territory as possible, so they do.

Not always. During the Dark Ages, Europe was fragmented and had few central governments. Partly because of armor. With armor, a sword, and a horse, you couldn't be stopped. Large states were hard to form because they rely on the threat of violence.

When gunpowder became popular in Europe, violence changed. In a world with guns, assembling large armies and conquest are cheaper.

James Dale Davidson says the internet will make nation-states obsolete. Most of the world's wealth will be online and in people's heads, making capital mobile.

Nation-states rely on predatory taxation of the rich to fund large militaries and welfare programs.

When capital is mobile, people can live anywhere in the world, Davidson says, making predatory taxation impossible. They're not bound by their job, land, or factory location. Wherever they're treated best.

Davidson says that over the next century, nation-states will collapse because they won't have enough money to operate as they do now. He imagines a world of small city-states, like Italy before 1900. (or Singapore today).

We've already seen some movement toward a more Sovereign Individual-like world. The pandemic proved large-scale remote work is possible, freeing workers from their location. Many cities and countries offer remote workers incentives to relocate.

Many Western businesspeople live in tax havens, and more people are renouncing their US citizenship due to high taxes. Increasing globalization has led to poor economic conditions and resentment among average people in the West, which is why politicians like Trump and Sanders rose to popularity with angry rhetoric, even though Obama rose to popularity with a more hopeful message.

The Sovereign Individual convinced me that the future will be different than Nassim Taleb's. Large countries like the U.S. will likely lose influence in the coming decades, while Portugal, Singapore, and Turkey will rise. If the trend toward less freedom continues, people may flee the West en masse.

So a traditional life of college, a big firm job, hard work, and corporate advancement may not be wise. Young people should learn as much as possible and develop flexible skills to adapt to the future.

Sapiens

Sapiens is a history of humanity, from proto-humans in Ethiopia to our internet society today, with some future speculation.

Sapiens views humans (and Homo sapiens) as a unique species on Earth. We were animals 100,000 years ago. We're slowly becoming gods, able to affect the climate, travel to every corner of the Earth (and the Moon), build weapons that can kill us all, and wipe out thousands of species.

Sapiens examines what makes Homo sapiens unique. Humans can believe in myths like religion, money, and human-made entities like countries and LLCs.

These myths facilitate large-scale cooperation. Ants from the same colony can cooperate. Any two humans can trade, though. Even if they're not genetically related, large groups can bond over religion and nationality.

Combine that with intelligence, and you have a species capable of amazing feats.

Sapiens may make your head explode because it looks at the world without presupposing values, unlike most books. It questions things that aren't usually questioned and says provocative things.

It also shows how human history works. It may help you understand and predict the world. Maybe.

The 4-hour Workweek

Things can be done better.

Tradition, laziness, bad bosses, or incentive structures cause complacency. If you're willing to make changes and not settle for the status quo, you can do whatever you do better and achieve more in less time.

The Four-Hour Work Week advocates this. Tim Ferriss explains how he made more sales in 2 hours than his 8-hour-a-day colleagues.

By firing 2 of his most annoying customers and empowering his customer service reps to make more decisions, he was able to leave his business and travel to Europe.

Ferriss shows how to escape your 9-to-5, outsource your life, develop a business that feeds you with little time, and go on mini-retirement adventures abroad.

Don't accept the status quo. Instead, level up. Find a way to improve your results. And try new things.

Why Nations Fail

Nogales, Arizona and Mexico were once one town. The US/Mexico border was arbitrarily drawn.

Both towns have similar cultures and populations. Nogales, Arizona is well-developed and has a high standard of living. Nogales, Mexico is underdeveloped and has a low standard of living. Whoa!

Why Nations Fail explains how government-created institutions affect country development. Strong property rights, capitalism, and non-corrupt governments promote development. Countries without capitalism, strong property rights, or corrupt governments don't develop.

Successful countries must also embrace creative destruction. They must offer ordinary citizens a way to improve their lot by creating value for others, not reducing them to slaves, serfs, or peasants. Authors say that ordinary people could get rich on trading expeditions in 11th-century Venice.

East and West Germany and North and South Korea have different economies because their citizens are motivated differently. It explains why Chile, China, and Singapore grow so quickly after becoming market economies.

People have spent a lot of money on third-world poverty. According to Why Nations Fail, education and infrastructure aren't the answer. Developing nations must adopt free-market economic policies.

Elon Musk

Elon Musk is the world's richest man, but that’s not a good way to describe him. Elon Musk is the world's richest man, which is like calling Steve Jobs a turtleneck-wearer or Benjamin Franklin a printer.

Elon Musk does cool sci-fi stuff to help humanity avoid existential threats.

Oil will run out. We've delayed this by developing better extraction methods. We only have so much nonrenewable oil.

Our society is doomed if it depends on oil. Elon Musk invested heavily in Tesla and SolarCity to speed the shift to renewable energy.

Musk worries about AI: we'll build machines smarter than us. We won't be able to stop these machines if something goes wrong, just like cows can't fight humans. Neuralink: we need to be smarter to compete with AI when the time comes.

If Earth becomes uninhabitable, we need a backup plan. Asteroid or nuclear war could strike Earth at any moment. We may not have much time to react if it happens in a few days. We must build a new civilization while times are good and resources are plentiful.

Short-term problems dominate our politics, but long-term issues are more important. Long-term problems can cause mass casualties and homelessness. Musk demonstrates how to think long-term.

The main reason people are impressed by Elon Musk, and why Ashlee Vances' biography influenced me so much, is that he does impossible things.

Electric cars were once considered unprofitable, but Tesla has made them mainstream. SpaceX is the world's largest private space company.

People lack imagination and dismiss ununderstood ideas as impossible. Humanity is about pushing limits. Don't worry if your dreams seem impossible. Try it.

Thanks for reading.

I recently wrote about getting VCs excited about Marketplace start-ups. SaaS founders became envious!

Understanding how people wire tens of millions is the only Series A hack I recommend.

Few people understand the intellectual process behind investing.

VC is risk management.

Series A-focused VCs must cover two risks.

1. Market risk

You need a large market to cross a threshold beyond which you can build defensibilities. Series A VCs underwrite market risk.

They must see you have reached product-market fit (PMF) in a large total addressable market (TAM).

2. Execution risk

When evaluating your growth engine's blitzscaling ability, execution risk arises.

When investors remove operational uncertainty, they profit.

Series A VCs like businesses with derisked revenue streams. Don't raise unless you have a predictable model, pipeline, and growth.

Please beat these 3 metrics before Series A:

Achieve $1.5m ARR in 12-24 months (Market risk)

Above 100% Net Dollar Retention. (Market danger)

Lead Velocity Rate supporting $10m ARR in 2–4 years (Execution risk)

Hit the 3 and you'll raise $10M in 4 months. Discussing 2/3 may take 6–7 months.

If none, don't bother raising and focus on becoming a capital-efficient business (Topics for other posts).

Let's examine these 3 metrics for the brave ones.

1. Lead Velocity Rate supporting €$10m ARR in 2 to 4 years

Last because it's the least discussed. LVR is the most reliable data when evaluating a growth engine, in my opinion.

SaaS allows you to see the future.

Monthly Sales and Sales Pipelines, two predictive KPIs, have poor data quality. Both are lagging indicators, and minor changes can cause huge modeling differences.

Analysts and Associates will trash your forecasts if they're based only on Monthly Sales and Sales Pipeline.

LVR, defined as month-over-month growth in qualified leads, is rock-solid. There's no lag. You can See The Future if you use Qualified Leads and a consistent formula and process to qualify them.

With this metric in your hand, scaling your company turns into an execution play on which VCs are able to perform calculations risk.

2. Above-100% Net Dollar Retention.

Net Dollar Retention is a better-known SaaS health metric than LVR.

Net Dollar Retention measures a SaaS company's ability to retain and upsell customers. Ask what $1 of net new customer spend will be worth in years n+1, n+2, etc.

Depending on the business model, SaaS businesses can increase their share of customers' wallets by increasing users, selling them more products in SaaS-enabled marketplaces, other add-ons, and renewing them at higher price tiers.

If a SaaS company's annualized Net Dollar Retention is less than 75%, there's a problem with the business.

Slack's ARR chart (below) shows how powerful Net Retention is. Layer chart shows how existing customer revenue grows. Slack's S1 shows 171% Net Dollar Retention for 2017–2019.

Slack S-1

3. $1.5m ARR in the last 12-24 months.

According to Point 9, $0.5m-4m in ARR is needed to raise a $5–12m Series A round.

Target at least what you raised in Pre-Seed/Seed. If you've raised $1.5m since launch, don't raise before $1.5m ARR.

Capital efficiency has returned since Covid19. After raising $2m since inception, it's harder to raise $1m in ARR.

P9's 2016-2021 SaaS Funding Napkin

In summary, less than 1% of companies VCs meet get funded. These metrics can help you win.

If there’s demand for it, I’ll do one on direct-to-consumer.

Cheers!

In this essay, I studied REAL NFT use examples and their potential uses.

Knowledge of the Hype Cycle

Gartner's Hype Cycle.

It proposes 5 phases for disruptive technology.

1. Technology Trigger: the emergence of potentially disruptive technology.

2. Peak of Inflated Expectations: Early publicity creates hype. (Ex: 2021 Bubble)

3. Trough of Disillusionment: Early projects fail to deliver on promises and the public loses interest. I suspect NFTs are somewhere around this trough of disillusionment now.

4. Enlightenment slope: The tech shows successful use cases.

5. Plateau of Productivity: Mainstream adoption has arrived and broader market applications have proven themselves. Here’s a more detailed visual of the Gartner Hype Cycle from Wikipedia.

In the speculative NFT bubble of 2021, @beeple sold Everydays: the First 5000 Days for $69 MILLION in 2021's NFT bubble.

@nbatopshot sold millions in video collectibles.

This is when expectations peaked.

Let's examine NFTs' real-world applications.

Watch this video if you're unfamiliar with NFTs.

Online Art

Most people think NFTs are rich people buying worthless JPEGs and MP4s.

Digital artwork and collectibles are revolutionary for creators and enthusiasts.

NFT Profile Pictures

You might also have seen NFT profile pictures on Twitter.

My profile picture is an NFT I coined with @skogards factoria app, which helps me avoid bogus accounts.

Profile pictures are a good beginning point because they're unique and clearly yours.

NFTs are a way to represent proof-of-ownership. It’s easier to prove ownership of digital assets than physical assets, which is why artwork and pfps are the first use cases.

They can do much more.

NFTs can represent anything with a unique owner and digital ownership certificate. Domains and usernames.

Usernames & Domains

@unstoppableweb, @ensdomains, @rarible sell NFT domains.

NFT domains are transferable, which is a benefit.

Godaddy and other web2 providers have difficult-to-transfer domains. Domains are often leased instead of purchased.

Tickets

NFTs can also represent concert tickets and event passes.

There's a limited number, and entry requires proof.

NFTs can eliminate the problem of forgery and make it easy to verify authenticity and ownership.

NFT tickets can be traded on the secondary market, which allows for:

1. marketplaces that are uniform and offer the seller and buyer security (currently, tickets are traded on inefficient markets like FB & craigslist)

2. unbiased pricing

3. Payment of royalties to the creator

4. Historical ticket ownership data implies performers can airdrop future passes, discounts, etc.

5. NFT passes can be a fandom badge.

The $30B+ online tickets business is increasing fast.

NFT-based ticketing projects:

• @GetProtocol

• @SeatlabNFT

• @YellowHeartNFT

• @GUTStickets

• @WicketEvents

Gaming Assets

NFTs also help in-game assets.

Imagine someone spending five years collecting a rare in-game blade, then outgrowing or quitting the game. Gamers value that collectible.

The gaming industry is expected to make $200 BILLION in revenue this year, a significant portion of which comes from in-game purchases.

Royalties on secondary market trading of gaming assets encourage gaming businesses to develop NFT-based ecosystems.

Digital assets are the start. On-chain NFTs can represent real-world assets effectively.

Real estate has a unique owner and requires ownership confirmation.

Real Estate

Tokenizing property has many benefits.

1. Can be fractionalized to increase access, liquidity

2. Can be collateralized to increase capital efficiency and access to loans backed by an on-chain asset

3. Allows investors to diversify or make bets on specific neighborhoods, towns or cities +++

I've written about this thought exercise before.

I made an animated video explaining this.

We've just explored NFTs for transferable assets. But what about non-transferrable NFTs?

SBTs are Soul-Bound Tokens. Vitalik Buterin (Ethereum co-founder) blogged about this.

NFTs are basically verifiable digital certificates.

Diplomas & Degrees

That fits Degrees & Diplomas. These shouldn't be marketable, thus they can be non-transferable SBTs.

Anyone can verify the legitimacy of on-chain credentials, degrees, abilities, and achievements.

The same goes for other awards.

For example, LinkedIn could give you a verified checkmark for your degree or skills.

Authenticity Protection

NFTs can also safeguard against counterfeiting.

Counterfeiting is the largest criminal enterprise in the world, estimated to be $2 TRILLION a year and growing.

Anti-counterfeit tech is valuable.

This is one of @ORIGYNTech's projects.

Identity

Identity theft/verification is another real-world problem NFTs can handle.

In the US, 15 million+ citizens face identity theft every year, suffering damages of over $50 billion a year.

This isn't surprising considering all you need for US identity theft is a 9-digit number handed around in emails, documents, on the phone, etc.

Identity NFTs can fix this.

• NFTs are one-of-a-kind and unforgeable.

• NFTs offer a universal standard.

• NFTs are simple to verify.

• SBTs, or non-transferrable NFTs, are tied to a particular wallet.

• In the event of wallet loss or theft, NFTs may be revoked.

This could be one of the biggest use cases for NFTs.

Imagine a global identity standard that is standardized across countries, cannot be forged or stolen, is digital, easy to verify, and protects your private details.

Since your identity is more than your government ID, you may have many NFTs.

@0xPolygon and @civickey are developing on-chain identity.

Memberships

NFTs can authenticate digital and physical memberships.

Voting

NFT IDs can verify votes.

If you remember 2020, you'll know why this is an issue.

Online voting's ease can boost turnout.

Informational property

NFTs can protect IP.

This can earn creators royalties.

NFTs have 2 important properties:

• Verifiability IP ownership is unambiguously stated and publicly verified.

• Platforms that enable authors to receive royalties on their IP can enter the market thanks to standardization.

Content Rights

Monetization without copyrighting = more opportunities for everyone.

This works well with the music.

Spotify and Apple Music pay creators very little.

Crowdfunding

Creators can crowdfund with NFTs.

NFTs can represent future royalties for investors.

This is particularly useful for fields where people who are not in the top 1% can’t make money. (Example: Professional sports players)

Mirror.xyz allows blog-based crowdfunding.

Financial NFTs

This introduces Financial NFTs (fNFTs). Unique financial contracts abound.

Examples:

• a person's collection of assets (unique portfolio)

• A loan contract that has been partially repaid with a lender

• temporal tokens (ex: veCRV)

Legal Agreements

Not just financial contracts.

NFT can represent any legal contract or document.

Messages & Emails

What about other agreements? Verbal agreements through emails and messages are likewise unique, but they're easily lost and fabricated.

Health Records

Medical records or prescriptions are another types of documentation that has to be verified but isn't.

Medical NFT examples:

• Immunization records

• Covid test outcomes

• Prescriptions

• health issues that may affect one's identity

• Observations made via health sensors

Existing systems of proof by paper / PDF have photoshop-risk.

I tried to include most use scenarios, but this is just the beginning.

NFTs have many innovative uses.

For example: @ShaanVP minted an NFT called “5 Minutes of Fame” 👇

Here are 2 Twitter threads about NFTs:

1. This piece of gold by @chriscantino

2. This conversation between @punk6529 and @RaoulGMI on @RealVision“The World According to @punk6529”

If you're wondering why NFTs are better than web2 databases for these use scenarios, see this Twitter thread I wrote:

If you liked this, please share it.

Deep learning-based generative art is being researched. As usual, self-learning is better. Some models, like OpenAI's DALL-E 2, require registration and can only be used online, but others can be used locally, which is usually more enjoyable for curious users. I'll demonstrate the Stable Diffusion model's operation on a standard PC.

Let’s get started.

What It Does

Stable Diffusion uses numerous components:

• A generative model trained to produce images is called a diffusion model. The model is incrementally improving the starting data, which is only random noise. The model has an image, and while it is being trained, the reversed process is being used to add noise to the image. Being able to reverse this procedure and create images from noise is where the true magic is (more details and samples can be found in the paper).

• An internal compressed representation of a latent diffusion model, which may be altered to produce the desired images, is used (more details can be found in the paper). The capacity to fine-tune the generation process is essential because producing pictures at random is not very attractive (as we can see, for instance, in Generative Adversarial Networks).

• A neural network model called CLIP (Contrastive Language-Image Pre-training) is used to translate natural language prompts into vector representations. This model, which was trained on 400,000,000 image-text pairs, enables the transformation of a text prompt into a latent space for the diffusion model in the scenario of stable diffusion (more details in that paper).

This figure shows all data flow:

The weights file size for Stable Diffusion model v1 is 4 GB and v2 is 5 GB, making the model quite huge. The v1 model was trained on 256x256 and 512x512 LAION-5B pictures on a 4,000 GPU cluster using over 150.000 NVIDIA A100 GPU hours. The open-source pre-trained model is helpful for us. And we will.

Install

Before utilizing the Python sources for Stable Diffusion v1 on GitHub, we must install Miniconda (assuming Git and Python are already installed):

wget https://repo.anaconda.com/miniconda/Miniconda3-py39_4.12.0-Linux-x86_64.sh
chmod +x Miniconda3-py39_4.12.0-Linux-x86_64.sh
./Miniconda3-py39_4.12.0-Linux-x86_64.sh
conda update -n base -c defaults conda

Install the source and prepare the environment:

git clone https://github.com/CompVis/stable-diffusion
cd stable-diffusion
conda env create -f environment.yaml
conda activate ldm
pip3 install transformers --upgrade

Download the pre-trained model weights next. HiggingFace has the newest checkpoint sd-v14.ckpt (a download is free but registration is required). Put the file in the project folder and have fun:

python3 scripts/txt2img.py --prompt "hello world" --plms --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

Almost. The installation is complete for happy users of current GPUs with 12 GB or more VRAM. RuntimeError: CUDA out of memory will occur otherwise. Two solutions exist.

Running the optimized version

Try optimizing first. After cloning the repository and enabling the environment (as previously), we can run the command:

python3 optimizedSD/optimized_txt2img.py --prompt "hello world" --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

Stable Diffusion worked on my visual card with 8 GB RAM (alas, I did not behave well enough to get NVIDIA A100 for Christmas, so 8 GB GPU is the maximum I have;).

Running Stable Diffusion without GPU

If the GPU does not have enough RAM or is not CUDA-compatible, running the code on a CPU will be 20x slower but better than nothing. This unauthorized CPU-only branch from GitHub is easiest to obtain. We may easily edit the source code to use the latest version. It's strange that a pull request for that was made six months ago and still hasn't been approved, as the changes are simple. Readers can finish in 5 minutes:

• Replace if attr.device!= torch.device(cuda) with if attr.device!= torch.device(cuda) and torch.cuda.is available at line 20 of ldm/models/diffusion/ddim.py ().

• Replace if attr.device!= torch.device(cuda) with if attr.device!= torch.device(cuda) and torch.cuda.is available in line 20 of ldm/models/diffusion/plms.py ().

• Replace device=cuda in lines 38, 55, 83, and 142 of ldm/modules/encoders/modules.py with device=cuda if torch.cuda.is available(), otherwise cpu.

• Replace model.cuda() in scripts/txt2img.py line 28 and scripts/img2img.py line 43 with if torch.cuda.is available(): model.cuda ().

Run the script again.

Testing

Test the model. Text-to-image is the first choice. Test the command line example again:

python3 scripts/txt2img.py --prompt "hello world" --plms --ckpt sd-v1-4.ckpt --skip_grid --n_samples 1

The slow generation takes 10 seconds on a GPU and 10 minutes on a CPU. Final image:

Hello world is dull and abstract. Try a brush-wielding hamster. Why? Because we can, and it's not as insane as Napoleon's cat. Another image:

Generating an image from a text prompt and another image is interesting. I made this picture in two minutes using the image editor (sorry, drawing wasn't my strong suit):

I can create an image from this drawing:

python3 scripts/img2img.py --prompt "A bird is sitting on a tree branch" --ckpt sd-v1-4.ckpt --init-img bird.png --strength 0.8

It was far better than my initial drawing:

I hope readers understand and experiment.

Stable Diffusion UI

Developers love the command line, but regular users may struggle. Stable Diffusion UI projects simplify image generation and installation. Simple usage:

• Unpack the ZIP after downloading it from https://github.com/cmdr2/stable-diffusion-ui/releases. Linux and Windows are compatible with Stable Diffusion UI (sorry for Mac users, but those machines are not well-suitable for heavy machine learning tasks anyway;).

• Start the script.

Done. The web browser UI makes configuring various Stable Diffusion features (upscaling, filtering, etc.) easy:

V2.1 of Stable Diffusion

I noticed the notification about releasing version 2.1 while writing this essay, and it was intriguing to test it. First, compare version 2 to version 1:

• alternative text encoding. The Contrastive LanguageImage Pre-training (CLIP) deep learning model, which was trained on a significant number of text-image pairs, is used in Stable Diffusion 1. The open-source CLIP implementation used in Stable Diffusion 2 is called OpenCLIP. It is difficult to determine whether there have been any technical advancements or if legal concerns were the main focus. However, because the training datasets for the two text encoders were different, the output results from V1 and V2 will differ for the identical text prompts.

• a new depth model that may be used to the output of image-to-image generation.

• a revolutionary upscaling technique that can quadruple the resolution of an image.

• Generally higher resolution Stable Diffusion 2 has the ability to produce both 512x512 and 768x768 pictures.

The Hugging Face website offers a free online demo of Stable Diffusion 2.1 for code testing. The process is the same as for version 1.4. Download a fresh version and activate the environment:

conda deactivate  
conda env remove -n ldm  # Use this if version 1 was previously installed
git clone https://github.com/Stability-AI/stablediffusion
cd stablediffusion
conda env create -f environment.yaml
conda activate ldm

Hugging Face offers a new weights ckpt file.

The Out of memory error prevented me from running this version on my 8 GB GPU. Version 2.1 fails on CPUs with the slow conv2d cpu not implemented for Half error (according to this GitHub issue, the CPU support for this algorithm and data type will not be added). The model can be modified from half to full precision (float16 instead of float32), however it doesn't make sense since v1 runs up to 10 minutes on the CPU and v2.1 should be much slower. The online demo results are visible. The same hamster painting with a brush prompt yielded this result:

It looks different from v1, but it functions and has a higher resolution.

The superresolution.py script can run the 4x Stable Diffusion upscaler locally (the x4-upscaler-ema.ckpt weights file should be in the same folder):

python3 scripts/gradio/superresolution.py configs/stable-diffusion/x4-upscaling.yaml x4-upscaler-ema.ckpt

This code allows the web browser UI to select the image to upscale:

The copy-paste strategy may explain why the upscaler needs a text prompt (and the Hugging Face code snippet does not have any text input as well). I got a GPU out of memory error again, although CUDA can be disabled like v1. However, processing an image for more than two hours is unlikely:

Stable Diffusion Limitations

When we use the model, it's fun to see what it can and can't do. Generative models produce abstract visuals but not photorealistic ones. This fundamentally limits The generative neural network was trained on text and image pairs, but humans have a lot of background knowledge about the world. The neural network model knows nothing. If someone asks me to draw a Chinese text, I can draw something that looks like Chinese but is actually gibberish because I never learnt it. Generative AI does too! Humans can learn new languages, but the Stable Diffusion AI model includes only language and image decoder brain components. For instance, the Stable Diffusion model will pull NO WAR banner-bearers like this:

V1:

V2.1:

The shot shows text, although the model never learned to read or write. The model's string tokenizer automatically converts letters to lowercase before generating the image, so typing NO WAR banner or no war banner is the same.

I can also ask the model to draw a gorgeous woman:

V1:

V2.1:

The first image is gorgeous but physically incorrect. A second one is better, although it has an Uncanny valley feel. BTW, v2 has a lifehack to add a negative prompt and define what we don't want on the image. Readers might try adding horrible anatomy to the gorgeous woman request.

If we ask for a cartoon attractive woman, the results are nice, but accuracy doesn't matter:

V1:

V2.1:

Another example: I ordered a model to sketch a mouse, which looks beautiful but has too many legs, ears, and fingers:

V1:

V2.1: improved but not perfect.

V1 produces a fun cartoon flying mouse if I want something more abstract:

I tried multiple times with V2.1 but only received this:

The image is OK, but the first version is closer to the request.

Stable Diffusion struggles to draw letters, fingers, etc. However, abstract images yield interesting outcomes. A rural landscape with a modern metropolis in the background turned out well:

V1:

V2.1:

Generative models help make paintings too (at least, abstract ones). I searched Google Image Search for modern art painting to see works by real artists, and this was the first image:

I typed "abstract oil painting of people dancing" and got this:

V1:

V2.1:

It's a different style, but I don't think the AI-generated graphics are worse than the human-drawn ones.

The AI model cannot think like humans. It thinks nothing. A stable diffusion model is a billion-parameter matrix trained on millions of text-image pairs. I input "robot is creating a picture with a pen" to create an image for this post. Humans understand requests immediately. I tried Stable Diffusion multiple times and got this:

This great artwork has a pen, robot, and sketch, however it was not asked. Maybe it was because the tokenizer deleted is and a words from a statement, but I tried other requests such robot painting picture with pen without success. It's harder to prompt a model than a person.

I hope Stable Diffusion's general effects are evident. Despite its limitations, it can produce beautiful photographs in some settings. Readers who want to use Stable Diffusion results should be warned. Source code examination demonstrates that Stable Diffusion images feature a concealed watermark (text StableDiffusionV1 and SDV2) encoded using the invisible-watermark Python package. It's not a secret, because the official Stable Diffusion repository's test watermark.py file contains a decoding snippet. The put watermark line in the txt2img.py source code can be removed if desired. I didn't discover this watermark on photographs made by the online Hugging Face demo. Maybe I did something incorrectly (but maybe they are just not using the txt2img script on their backend at all).

Conclusion

The Stable Diffusion model was fascinating. As I mentioned before, trying something yourself is always better than taking someone else's word, so I encourage readers to do the same (including this article as well;).

Is Generative AI a game-changer? My humble experience tells me:

• I think that place has a lot of potential. For designers and artists, generative AI can be a truly useful and innovative tool. Unfortunately, it can also pose a threat to some of them since if users can enter a text field to obtain a picture or a website logo in a matter of clicks, why would they pay more to a different party? Is it possible right now? unquestionably not yet. Images still have a very poor quality and are erroneous in minute details. And after viewing the image of the stunning woman above, models and fashion photographers may also unwind because it is highly unlikely that AI will replace them in the upcoming years.

• Today, generative AI is still in its infancy. Even 768x768 images are considered to be of a high resolution when using neural networks, which are computationally highly expensive. There isn't an AI model that can generate high-resolution photographs natively without upscaling or other methods, at least not as of the time this article was written, but it will happen eventually.

• It is still a challenge to accurately represent knowledge in neural networks (information like how many legs a cat has or the year Napoleon was born). Consequently, AI models struggle to create photorealistic photos, at least where little details are important (on the other side, when I searched Google for modern art paintings, the results are often even worse;).

• When compared to the carefully chosen images from official web pages or YouTube reviews, the average output quality of a Stable Diffusion generation process is actually less attractive because to its high degree of randomness. When using the same technique on their own, consumers will theoretically only view those images as 1% of the results.

Anyway, it's exciting to witness this area's advancement, especially because the project is open source. Google's Imagen and DALL-E 2 can also produce remarkable findings. It will be interesting to see how they progress.