Bilingual

What Made Lisp Different
a day ago
- Conditionals (if-then-else) were invented by McCarthy for Lisp and later spread to most languages.
- Functions are first-class objects in Lisp, treated as a data type like integers or strings.
- Recursion was first supported as a programming language feature in Lisp.
- Variables in Lisp are effectively pointers; values have types, not variables.
- Lisp introduced garbage-collection for automatic memory management.
- Programs are composed of expressions that each return a value, unlike the expression/statement distinction in Fortran.
- Lisp has a symbol type distinct from strings, allowing equality tests via pointer comparison.
- Code is represented using trees of symbols, enabling powerful macros (still unique to Lisp).
- The whole language is always available: no distinction between read-time, compile-time, and runtime, allowing syntax reprogramming and macros.
The Word "Hacker"
a day ago
- The term 'hacker' has two meanings: a computer intruder in popular press, and a skilled programmer among programmers.
- The noun 'hack' can refer to an ugly solution or a clever rule-breaking solution, with ugly ones being more common.
- Hacking predates computers, as seen in Richard Feynman's safe-cracking during the Manhattan Project.
- Hackers' disobedience stems from intellectual curiosity, which is often misunderstood by authorities.
- Hackers oppose strong intellectual property laws because they threaten the freedom needed to innovate.
- Major tech innovations (e.g., IBM's business computer, Unix) often come from outsiders, not established groups.
- New intellectual property laws restrict studying current technology, hindering development of new ideas.
- Hackers are uniformly critical of these laws, suggesting they may harm rather than protect America.
- Hacking's unruly essence aligns with American values, fostering innovation and wealth, as seen in Silicon Valley.
- Civil liberties are economically beneficial, and hackers are sensitive to their erosion, which could lead to bad ideas winning.
What the Bubble Got Right
a day ago
- The Internet Bubble was driven by a Ponzi-like cycle where startups invested ad revenue into Yahoo, inflating its earnings and stock price, but Yahoo remained fundamentally valuable after the crash.
- Taking companies public early (retail VC) is not inherently flawed; markets can learn to value pre-earnings companies, and going public early can be beneficial for some startups.
- The Internet is a transformative technology, but most profits from it come indirectly—like railroads benefiting steel and oil companies—rather than from pure 'Internet companies.'
- The Internet reduces bottlenecks by giving consumers more choices, making it easier for quality products to succeed through word-of-mouth and low-cost discovery.
- Youthful startup founders are a lasting trend; while they may lack experience in management, their vision and technical insight are increasingly critical to success.
- Informality (e.g., open-necked shirts) reflects a meritocracy where substance outweighs presentation; formality often masks a lack of good ideas.
- Nerds focus on substance over marketing, and as technology grows in importance, nerd culture rises, driven by the tangible success of technical innovation.
- Stock options are fair and effective for motivating employees, but they can incentivize short-term stock pumping; tweaks tying them to earnings may improve alignment.
- Startups are increasingly designed to be acquired, offering a low-risk way for larger companies to acquire technology developed in innovative, agile environments.
- Silicon Valley remains a hub of progressive, future-focused energy and productivity, despite the bubble's aftermath, and is poised to generate new wealth.
- Technology amplifies productivity variations, enabling very small companies (e.g., under 10 people) to achieve large revenues by focusing on product development and outsourcing.
- Overall, the future favors good ideas over connections or appearances, accelerating the rate at which value creation is rewarded—a true 'new economy.'
Made in USA
a day ago
- American impatience leads to success in fields like software and movies, where rapid prototyping and boldness are beneficial.
- The same impatience causes failure in physical products like cars and cities, where good design and craftsmanship are crucial.
- U.S. car and housing design is often driven by marketing rather than genuine design principles, resulting in poor quality.
- Japanese culture's emphasis on design and craftsmanship explains their superior products, from swords to cars.
- Apple stands out as an exception in the U.S., combining American speed with meticulous design under CEO leadership.
- Speed and innovation in the U.S. generate wealth, but there is potential to balance impatience with good design.
- Future success may come from ignoring national stereotypes and adopting best practices for each industry.
Ideas for Startups
a day ago
- Startup ideas are not million-dollar ideas; they are starting points that evolve through exploration and questioning.
- The best environment for generating ideas involves exposure to new technology and collaboration with like-minded friends, such as in a university.
- Ideas often arise from mental habits and doodling, requiring a real problem to focus the mind.
- To create a successful startup, make something people want by fixing broken products, commoditizing luxuries, or simplifying usage.
- Most startups are acquired, so consider building products that multiple large companies might want, and let ideas emerge naturally from fun projects with friends.
Are Software Patents Evil?
a day ago
- The author was granted four patents despite applying for three, highlighting the prevalence of software patents.
- Software patents are fundamentally no different from hardware patents, as modern machines rely increasingly on software.
- The patent office's inability to evaluate software novelty leads to granting obvious patents, like Amazon's one-click patent.
- Startups should not worry about infringing patents; they rarely get sued, and focus should be on building great products.
- Patents serve primarily as a defensive tool for startups seeking acquisition or growth, aiding in the 'mating dance' with acquirers.
- Patent trolls exploit the system by suing successful companies for profit, but they do not significantly hinder innovation.
- Overall, software patents have a minor impact on innovation, with startups ignoring them and focusing on user growth.
The Future of Web Startups
a day ago
- Startups are becoming cheaper to start, leading to a surge in new companies, similar to how technology becomes a commodity.
- Standardization of investment terms and acquisition processes will increase as startup volumes grow.
- Younger and more technical founders will be able to start startups without needing investor approval first.
- Startup hubs like Silicon Valley will persist and become even more important despite lower startup costs.
- Investors and acquirers will need better judgement to handle the increasing number of startups.
- College education will shift focus from degrees to practical skills and co-founder connections.
- More competitors will emerge, but the overall wealth creation is not zero-sum; technology will advance faster.
- Big companies are less productive for innovation; startups will drive faster technological evolution.
Why There Aren't More Googles
a day ago
- Startups that reject acquisition offers often achieve greater success, with larger offers or IPOs later.
- Google and Facebook stayed independent because acquirers underestimated their value, not just idealism.
- Venture capitalists are overly conservative, driven by consensus, and miss truly bold innovative ideas.
- There is a funding gap between early seed rounds (e.g., $20k) and large VC rounds (e.g., $2M), leaving promising startups needing ~$200k underserved.
- The declining cost of startups creates an opportunity for new investors to fill this gap by making smaller, bolder investments.
- VC firms resist adapting to smaller investments, but evolving models could produce more breakthrough companies like Google.
Frighteningly Ambitious Startup Ideas
a day ago
- Ambitious startup ideas often feel frightening because they challenge your identity and ambition.
- A new search engine could compete with Google by focusing on what hackers want, such as fast, clutter-free results.
- Replace email with a todo list protocol that gives recipients more control and is faster than Gmail.
- Universities may be partially replaced by diverse learning methods, separating credentials from education.
- Deliver internet drama via platforms like Netflix or through payment/streaming infrastructure companies.
- The next hardware visionary could come from a startup, following the example set by Steve Jobs and Apple.
- Revive Moore's Law through software that simplifies parallel computing, potentially via a 'sufficiently smart compiler' or a marketplace for optimization.
- Ongoing automatic medical diagnosis could replace reactive healthcare, despite resistance from medical traditions and false-alarm concerns.
- Tactically, start with small, seemingly harmless projects and expand westward rather than attacking big problems directly.
How to Get Startup Ideas
a day ago
- The best startup ideas come from problems founders themselves have, not from trying to think up ideas.
- Successful startups often begin with a small, urgent need (a 'well') rather than a broad but shallow interest.
- Founders should 'live in the future' and build what seems missing, often by noticing gaps in their own experience.
- Avoid 'made-up' or 'sitcom' ideas; focus on real problems that users urgently need solved.
- Being at the leading edge of a changing field helps you notice opportunities that others miss.
- Turn off filters like 'unsexy' and 'schlep' to see valuable ideas that others avoid.
- If you need an idea quickly, look for unmet needs in your own life or others', or seek out dying industries ripe for replacement.
How to Be an Expert in a Changing World
a day ago
- Confidence in beliefs can grow monotonically only in a static world; change undermines that certainty.
- Experts often fail because they rely on outdated knowledge about a changed world.
- Successful startup investors recognize that good ideas often seem bad initially due to recent changes.
- To avoid obsolete beliefs, explicitly acknowledge change and focus on being open-minded rather than predicting the future.
- New ideas arise from solving problems while valuing weird hunches, and betting on people (earnest, energetic, independent) over ideas is more reliable.
- Publicly betting on ideas, as investors do, creates strong incentives to correct outdated views.
- Accelerating change increases the challenge of not becoming a prisoner of one's own expertise.
What Microsoft Is this the Altair Basic of?
a day ago
- Successful startups often seem lame or wrong when they first launch, like stepping into a swamp rather than climbing a mountain.
- Founders are drawn to promising ideas by instinct because they live in the future and sense something missing, even if they can't articulate why.
- Instead of ridiculing lame-sounding startup ideas, ask 'What Microsoft is this the Altair Basic of?' to uncover their potential.
- Sometimes there are multiple possible 'Microsofts' a startup could grow into, varying in size by orders of magnitude.
Novelty and Heresy
a day ago
- Discovering new things often leads to accusations of heresy against established assumptions.
- Good but non-obvious ideas are hidden by mistaken assumptions that people strongly defend.
- Resisting heresy creates 'dead zones' of unexplored ideas around cherished false beliefs.
- Suppressing heresies blocks both direct contradictions and indirect implications against assumptions.
- Actively seeking heresies can turn dead zones into valuable sources for new ideas.
Fashionable Problems
a day ago
- Many fields have only a small fraction of possibilities explored, despite extensive work.
- Even the smartest people tend to work on fashionable problems, being conservative in their choices.
- Unfashionable problems can be found in fields thought to be fully explored, like essays, Lisp, or venture funding.
- Discovering a new approach in a big but played-out field multiplies the value of the discovery.
- Genuinely loving what you do is the best protection against working on the same things as everyone else.
Name It, and They Will Come — overreacted
a day ago
- The author describes the journey of discovering a new idea, building it into a project, and preparing for its public launch.
- Upon release, the project receives mostly irrelevant feedback, focusing on coding style, design, or generic industry debates rather than the core idea.
- The author suggests that truly novel ideas lack a shared vocabulary, causing people to fall back on discussing familiar topics like formatting or industry gossip.
- To overcome this, the author advises naming the problem and telling a story to make the new idea understandable and discussable.
<antirez>
a day ago
- Complexity is being ignored when adding features or optimizing, harming software quality.
- Complex build systems and absurd dependency chains make software bloated and fragile.
- Telling new programmers not to 'reinvent the wheel' hinders learning and innovation.
- Lack of backward API compatibility and unnecessary rewrites damage existing systems.
- Adopting every new language, paradigm, or framework without consideration is destructive.
- Underestimating the difficulty of working with existing libraries leads to poor choices.
- Assuming de facto standards are always better than tailored solutions is harmful.
- Disregarding code comments and viewing software as purely engineering is misguided.
- Systems that cannot scale down make simple tasks unnecessarily complex.
- Prioritizing speed over thoughtful design destroys the joy of hacking.
You don't hire juniors to do menial chores
2 days ago
- The notion that LLMs replace juniors is flawed because juniors are essential for future senior succession.
- Juniors are a cost-effective source of innovation, bringing fresh knowledge and challenging existing processes.
- Hiring juniors acts as a stress test, improving organizational processes by exposing gaps and assumptions.
- Relying on LLMs long-term may be more expensive due to potential price hikes and vendor lock-in, while junior development is cheaper and builds cultural ambassadors.
Private healthcare makes industries less innovative. It's time for change
5 days ago
- Nearly 1 in 4 American workers stay in unwanted jobs due to employer-based health insurance, a figure rising to 41% for those with chronic conditions.
- Private healthcare suppresses wages and working conditions by reducing workers' mobility and employers' incentive to improve.
- Job lock prevents innovators from leaving to pursue new ideas, stifling industry innovation, especially in sectors like news.
- Universal healthcare, like the UK's NHS, removes fear of medical bills, improves health, and promotes innovation by enabling career mobility.
- Even libertarian think tanks acknowledge employer-sponsored health insurance creates coverage gaps and reduces income mobility.
Stop pretending billionaires built the future
6 days ago
- High taxes on the rich in mid-20th century America led to greater economic equality and innovation, such as the GI Bill and NASA funding.
- Tax cuts since the 1980s have increased wealth inequality and stagnant wages, linking innovation primarily to billionaire incentives.
- Historical evidence shows innovation thrived during periods of high taxation and government investment, not billionaire payouts.
Painting the sides of railroad rails white to reduce derailment
11 days ago
- Union Pacific painted rails white in high-heat areas to reflect sunlight and reduce rail surface temperature by about 20 degrees, lowering the risk of thermal misalignments.
- This innovative method blends European rail techniques with U.S. highway striping practices, applied using a high-rail truck and paint sprayer.
- The approach contributed to a broader safety strategy, helping Union Pacific achieve its best-ever full-year derailment incident rate in 2025, a 19% year-over-year improvement.
- The initiative represents a culture of safety innovation, combining proven practices, continuous monitoring, and practical thinking to strengthen rail safety.
- Rail is consistently the safest way to move freight over land, with incident rates far lower than trucking when measured by gross ton-miles.

innovation

What Made Lisp Different

The Word "Hacker"

What the Bubble Got Right

Made in USA

Ideas for Startups

Are Software Patents Evil?

The Future of Web Startups

Why There Aren't More Googles

Frighteningly Ambitious Startup Ideas

How to Get Startup Ideas

How to Be an Expert in a Changing World

What Microsoft Is this the Altair Basic of?

Novelty and Heresy

Fashionable Problems

Name It, and They Will Come — overreacted

<antirez>

You don't hire juniors to do menial chores

Private healthcare makes industries less innovative. It's time for change

Stop pretending billionaires built the future

Painting the sides of railroad rails white to reduce derailment