Torvalds Rejects Visionary Label, Cites Stubborn Labor for Linux, Git

The Mind Behind Linux | Linus Torvalds | TED — TED

TED

Linus Torvalds argues transformative technology emerges not from grand visionary roadmaps, but from stubborn engineers quietly resolving immediate, practical bottlenecks.

AI-assisted summary of The Mind Behind Linux | Linus Torvalds | TED — original source. Check the original for full context.

Source published: 2016-05-03

Key ideas

Linus Torvalds created Linux and Git—which now power major internet infrastructure and over 1.5 billion Android devices—primarily out of personal necessity and a preference for solitary problem-solving rather than grand strategic roadmaps. (0:13)
Git was engineered specifically to break the human bottleneck caused by Torvalds managing kernel contributions from approximately 1,000 developers every two to three months. (7:13)
Torvalds defines 'good taste' in software engineering as an intuitive architectural ability to recognize systemic patterns and eliminate edge cases entirely, rather than handling them with conditional branches. (14:18)
Torvalds rejects the tech industry's romanticization of visionary savants like Nikola Tesla, aligning his pragmatic, ground-level engineering philosophy with Thomas Edison's emphasis on perspiration and iterative execution. (16:24)

Detailed notes

## Background Published on YouTube by TED on May 3, 2016, this interview features Linux and Git creator Linus Torvalds in conversation with TED Curator Chris Anderson. The session explores Torvalds' unexpected journey in transforming technology twice, highlighting his distinctive personality traits and pragmatic engineering philosophy.

## TL;DR Linus Torvalds reflects on how a solitary hobby project coded in his bathrobe evolved into the foundational operating system of the modern internet and over 1.5 billion Android devices. He explains how bottlenecks in community collaboration compelled him to invent Git, and why open-source models succeed by allowing people with opposing temperaments to build together without needing mutual affection. Torvalds illustrates his concept of 'good taste' in software architecture by demonstrating how elegant pointer design eliminates awkward edge cases in basic code. Ultimately, he rejects the label of visionary futurist, aligning himself instead with Thomas Edison's relentless, unglamorous persistence in fixing immediate, practical problems.

## Atmosphere & style The conversation carries an understated, refreshingly self-effacing tone, deliberately punctured with dry wit that contrasts sharply with typical Silicon Valley grandiosity. Torvalds speaks with disarming bluntness about his social awkwardness, lack of design taste, and introverted isolation, eliciting warm chuckles and spontaneous applause from the audience. Anderson acts as an engaged, curious foil, attempting to draw out visionary narratives only for Torvalds to playfully ground the dialogue back into practical, unromantic reality.

## Bulletin - [00:13](https://www.youtube.com/watch?v=o8NPllzkFhE&t=13s) Linus Torvalds created Linux and Git—which now power major internet infrastructure and over 1.5 billion Android devices—primarily out of personal necessity and a preference for solitary problem-solving rather than grand strategic roadmaps. - [07:13](https://www.youtube.com/watch?v=o8NPllzkFhE&t=433s) Git was engineered specifically to break the human bottleneck caused by Torvalds managing kernel contributions from approximately 1,000 developers every two to three months. - [14:18](https://www.youtube.com/watch?v=o8NPllzkFhE&t=858s) Torvalds defines 'good taste' in software engineering as an intuitive architectural ability to recognize systemic patterns and eliminate edge cases entirely, rather than handling them with conditional branches. - [16:24](https://www.youtube.com/watch?v=o8NPllzkFhE&t=984s) Torvalds rejects the tech industry's romanticization of visionary savants like Nikola Tesla, aligning his pragmatic, ground-level engineering philosophy with Thomas Edison's emphasis on perspiration and iterative execution.

## Key takeaways - [00:13](https://www.youtube.com/watch?v=o8NPllzkFhE&t=13s) Torvalds intentionally keeps his home workspace silent and free of distractions, opting for non-stimulating environments over elaborate setups. - [02:13](https://www.youtube.com/watch?v=o8NPllzkFhE&t=133s) Linux originated at age 21 as an isolated, personal programming exercise rather than an intentional global movement. - [02:13](https://www.youtube.com/watch?v=o8NPllzkFhE&t=133s) The release of Linux was prompted by a desire for feedback rather than a commercial agenda, with free software licensing adopted after an introduction by a university peer. - [04:44](https://www.youtube.com/watch?v=o8NPllzkFhE&t=284s) Asynchronous email communication served as an essential operational buffer that allowed Torvalds to collaborate without face-to-face social friction. - [04:44](https://www.youtube.com/watch?v=o8NPllzkFhE&t=284s) The critical transition for Linux was expanding from solo development to a circle of 10 to 100 contributors; subsequent growth to millions was gradual and organic. - [07:13](https://www.youtube.com/watch?v=o8NPllzkFhE&t=433s) Git was developed as a distributed source control system to eliminate Torvalds as the single point of failure managing kernel patch influx. - [09:23](https://www.youtube.com/watch?v=o8NPllzkFhE&t=563s) Torvalds attributes his accomplishments to unrelenting persistence and stubbornness rather than innate, visionary brilliance. - [11:57](https://www.youtube.com/watch?v=o8NPllzkFhE&t=717s) Open source collaboration succeeds because contributors with divergent personalities, goals, and communication styles can work together without needing mutual affection. - [11:57](https://www.youtube.com/watch?v=o8NPllzkFhE&t=717s) Commercial involvement benefited Linux by handling user interfaces, distribution, and business functions that Torvalds had no interest in managing. - [14:18](https://www.youtube.com/watch?v=o8NPllzkFhE&t=858s) Good taste in programming is exemplified by redesigning data structures to eliminate conditional special cases in favor of unified execution flow. - [16:24](https://www.youtube.com/watch?v=o8NPllzkFhE&t=984s) Torvalds rejects long-term vision and futurism, choosing instead to focus pragmatically on immediate engineering obstacles. - [16:24](https://www.youtube.com/watch?v=o8NPllzkFhE&t=984s) Drawing a comparison between Nikola Tesla and Thomas Edison, Torvalds favors Edison's unglamorous, execution-focused philosophy. - [18:51](https://www.youtube.com/watch?v=o8NPllzkFhE&t=1131s) Torvalds expresses zero resentment toward corporations building multi-billion-dollar businesses on Linux, crediting open access with its global ubiquity. - [18:51](https://www.youtube.com/watch?v=o8NPllzkFhE&t=1131s) Open source methodologies work particularly well in software because code provides objective, binary pass/fail validation that civic and political systems lack.

## Detailed walkthrough ### [00:13](https://www.youtube.com/watch?v=o8NPllzkFhE&t=13s) The "Linux World Headquarters" and Working in Isolation

Chris Anderson opens the conversation by highlighting the staggering footprint of Linus Torvalds' work: the Linux kernel runs millions of computers, underpins much of the internet, and powers over 1.5 billion active Android devices. Anderson remarks that one might expect an empire of this magnitude to be directed from an imposing corporate headquarters, yet the actual workspace shown on screen is modest to the point of comedy. Torvalds describes his home office as utterly devoid of excitement, noting that the most talked-about feature in the photo—a walking treadmill desk—is something he has already abandoned. For Torvalds, the work environment must eliminate all external distractions; the walls are painted a muted light green, a shade he wryly points out is often found in mental institutions to keep patients calm.

Torvalds places an extreme premium on absolute silence. While he acknowledges that some Google engineers install miniature data centers in their residences, his only hardware requirement beyond raw performance is that his computer make no audible sound whatsoever. Torvalds explains: “I want to hear the cat purring, not the sound of the fans in the computer.” Working alone in quiet solitude, he finds peace when his cat curls up in his lap, undisturbed by the buzzing machinery of modern computing.

- Linux powers a vast portion of internet infrastructure and over 1.5 billion active Android devices. - Torvalds’ home office setup features light green walls chosen specifically for their non-stimulating, calming effect. - Torvalds discarded his walking treadmill desk and insists on completely silent hardware to avoid distracting fan noise.

---

[02:13](https://www.youtube.com/watch?v=o8NPllzkFhE&t=133s) The Accidental Genesis of Linux and Free Software

Anderson asks how an engineer working alone in such quiet circumstances managed to orchestrate a vast technological movement through open source. Torvalds responds candidly that his day-to-day routine has not changed: he still codes by himself at home, frequently in his bathrobe, dressing up only when a visiting photographer compels him to put on clothes. Linux itself was never envisioned as a sweeping global collaboration. Instead, it originated as one of many personal programming exercises Torvalds undertook strictly for himself, driven partly by a need for the end result, but primarily by pure enjoyment of programming.

When Torvalds first released the code 25 years earlier, the modern open-source development methodology did not exist in his thinking. His initial release was simply an impulse to share what he had built over six months of solitary labor and solicit feedback. At the University of Helsinki, a friend embedded in the free software community introduced him to preexisting software licenses. Although Torvalds initially worried that commercial entities might exploit and take advantage of his labor, he ultimately chose to adopt a free software license, concluding, “What the hell?”

- Torvalds began Linux at age 21 purely as a personal programming project with no collaborative master plan. - The software was made publicly available to show off six months of work and gather critiques, rather than to launch a movement. - A classmate at the University of Helsinki introduced Torvalds to free software licensing, overcoming his initial wariness about commercial exploitation.

---

[04:44](https://www.youtube.com/watch?v=o8NPllzkFhE&t=284s) Community Feedback and the Gradual Scale of Collaboration

The collaborative engine of Linux began not with patches of code, but with external feedback and ideas. For Torvalds, who had already spent roughly half his life programming in complete isolation, having outside observers inspect his source code and offer suggestions felt like a revelation. Although this influx inspired an initial appreciation for other contributors, Torvalds humorously clarifies that he remains deeply introverted. Linus Torvalds remarks: “I don't really love other people -- but I love computers, I love interacting with other people on email, because it kind of gives you that buffer.”

```mermaid flowchart TD accTitle: Evolution of Linux Development accDescr: Diagram illustrating Linus Torvalds' progression from solitary coding to a large distributed open-source community.

subgraph Phase1["Solitary Development"] A["Linus codes in isolation<br/>( bathrobe, silent office)"] --> B["Public release of source code<br/>( seeking feedback)"] end

subgraph Phase2["Early Collaboration"] B --> C["Contributors submit ideas & critiques"] C --> D["First code contributions<br/>( community of 10 to 100)"] end

subgraph Phase3["Massive Scale"] D --> E["Gradual expansion to thousands"] E --> F["~1,000 contributors per release<br/>( every 2-3 months)"] end ```

When Anderson asks whether there was an explosive moment where Torvalds realized Linux was becoming an industry-defining phenomenon, Torvalds downplays the idea of a sudden turning point. The true paradigm shift for him occurred when the project transitioned from absolute solitude to a group of 10 or 100 people. From that baseline, the growth to a community of millions was an incremental, organic evolution spread across many years rather than an overnight commercial takeoff.

- Torvalds started coding around age 10 or 11, working in complete isolation until Linux was published. - The earliest contributions consisted of conceptual ideas, bug reports, and design critiques before actual code patches arrived. - Email communication served as an ideal operational buffer, enabling Torvalds to coordinate technical discussions without face-to-face interaction. - The primary milestone for Torvalds was escaping isolation with a group of 10 to 100 contributors; subsequent growth was gradual.

---

[07:13](https://www.youtube.com/watch?v=o8NPllzkFhE&t=433s) Breaking the Bottleneck: The Invention of Git

As the Linux developer community ballooned toward 10,000 participants—with roughly 1,000 developers contributing patches to every single release cycle every two to three months—the sheer scale broke existing collaboration paradigms. Torvalds found himself serving as an unsustainable human bottleneck. Traditional source code management systems of the era proved unworkable; he detested CVS with a passion and refused to touch it, while alternative tools were widely disliked across the community.

To eliminate this operational impasse, Torvalds designed Git, a distributed source-code management system built specifically to alleviate his own workload. Anderson observes the irony: both Linux and Git, two foundational pillars of modern technology, emerged largely as unintended side effects of Torvalds' desire to avoid interacting with large crowds of people. Torvalds agrees with this assessment, explaining that every major project he has ever tackled was born strictly out of personal technical necessity.

Linus Torvalds explains: “So Git is my second big project, which was only created for me to maintain my first big project.”

- Every Linux kernel release incorporates contributions from approximately 1,000 developers every two to three months. - Torvalds adamantly refused to use CVS, the industry's predominant version control system at the time. - Git was created specifically to remove Torvalds as the single point of failure in managing incoming kernel contributions.

---

[09:23](https://www.youtube.com/watch?v=o8NPllzkFhE&t=563s) Stubbornness, Solitude, and the Geek Temperament

Anderson introduces a childhood photograph of Torvalds alongside his younger brother, with Torvalds intently manipulating a Rubik’s Cube. Torvalds confirms he was the archetypal nerd: absorbed in math, physics, and computer science while lacking natural social instincts. He recalls that his younger sister would routinely brief him before family gatherings, whispering relatives' names to him outside the door to ensure he could navigate the room.

When asked whether innate genius drove his success, Torvalds defers, pointing instead to a character trait his sister identified: absolute, unrelenting persistence. Unlike engineers who chase the next flashy concept, Torvalds fixates on a problem and refuses to let go. This stubbornness manifests in his career choices as well. While Silicon Valley culture prizes job-hopping to accelerate careers, Torvalds spent seven consecutive years at the exact same company after moving to California.

Linus Torvalds reflects: “Apparently, my sister said that my biggest exceptional quality was that I would not let go.”

- Torvalds self-identifies as a classic geek who preferred solving Rubik's Cubes to socializing. - His sister routinely coached him on social dynamics and relative identities prior to family gatherings. - Torvalds spent seven unbroken years at a single company in Silicon Valley, defying the regional norm of frequent job-hopping.

[11:57](https://www.youtube.com/watch?v=o8NPllzkFhE&t=717s) Acknowledging Myopia and Welcoming Different Strengths in Open Source

Returning to his admission that he is not a "people person," Linus Torvalds reflects candidly on his personal shortcomings when dealing with others. He confesses to being somewhat myopic regarding other people’s feelings, acknowledging that this blindness occasionally causes him to say things that inflict emotional hurt—an aspect of his demeanor that he openly states he is not proud of, eliciting spontaneous applause from the audience. He notes the irony that when critics tell him he ought to be nice, his response—suggesting that perhaps they are too nice and should instead be more aggressive—is promptly taken as further proof of his abrasive nature. Rather than offering excuses, Torvalds explains that people are simply wired differently, and open source thrives precisely because it enables individuals with clashing temperaments to collaborate effectively without requiring mutual affection.

Torvalds emphasizes that open source communities can tolerate fierce, heated arguments because participants are often driven by entirely different motivations and care about completely different aspects of a project. He revisits his early skepticism regarding commercial adoption, admitting that the commercial entities he once feared would exploit his software turned out to be "lovely, lovely people" who eagerly tackled the exact tasks he despised. While an open source ecosystem needs warm, communicative individuals who welcome newcomers into the fold, it equally depends on blunt engineers who care strictly about the underlying technology. Torvalds jokes about his total lack of graphical sensibility, noting that while others obsess over interface design, his own abilities are nonexistent: “I mean, if I was stranded on an island and the only way to get off that island was the make a pretty UI, I'd die there.”

- Torvalds observes that participants in open source do not need to like each other, nor do they even have to formally agree to disagree; progress happens simply because different contributors pursue distinct goals. - Commercial enterprises stepped in to supply resources, develop user-facing elements, and pursue distribution strategies that Torvalds had no personal interest in leading.

[14:18](https://www.youtube.com/watch?v=o8NPllzkFhE&t=858s) Good Taste in Code: Eliminating the Special Case

Chris Anderson steers the conversation toward another core attribute Torvalds mentioned in their preparatory discussions: having "taste" in code. Prompted by an audience survey where dozens of hands shoot up to confirm they have written software, Torvalds examines two code slides depicting the deletion of an existing entry from a singly-linked list. He points out that the conventional solution taught to beginner programmers relies on a conditional statement to check whether the targeted node is the first element in the list or one in the middle, because updating the root pointer versus updating a previous node's pointer requires distinct handling.

The second slide, representing what Torvalds considers "better taste," removes the conditional `if` branch entirely by reframing the pointer manipulation so that the boundary case and the standard case are resolved identically. Torvalds clarifies that while this introductory linked-list problem is trivial CS 101 code, the underlying concept is foundational: good taste represents an instinctive ability to perceive systemic patterns and eliminate edge cases so that exceptions become part of the normal flow. For Torvalds, identifying collaborators who possess this higher-level architectural instinct is paramount. As Linus Torvalds argues, “Good taste is about really seeing the big patterns and kind of instinctively knowing what's the right way to do things.”

```mermaid flowchart TD accTitle: Code Taste Comparison in Pointer Manipulation accDescr: Diagram contrasting conventional branching code with good taste pointer architecture that eliminates the special case.

subgraph Conventional["Conventional Approach (Branching)"] A["Node Removal Request"] --> B{"Is it the first entry?"} B -->|"Yes (Special Case)"| C["Update Head Pointer"] B -->|"No (Normal Case)"| D["Update Previous Node's Pointer"] end

subgraph GoodTaste["Good Taste Approach (Unified Pattern)"] E["Node Removal Request"] --> F["Indirect Pointer Target"] F --> G["Uniform Pointer Reassignment without Branching"] end ```

- Singly-linked list deletion conventionally creates two divergent branches: mutating the head pointer versus mutating the `next` pointer of a preceding element. - Good taste transforms an awkward special case into standard behavior, simplifying execution flow and eliminating structural fragility. - Torvalds considers micro-examples like list removal merely illustrative; true taste manifests in larger structural decisions and architectural intuition across complex codebases.

[16:24](https://www.youtube.com/watch?v=o8NPllzkFhE&t=984s) The Engineer versus the Visionary: Edison over Tesla

Anderson attempts to assemble Torvalds's traits—technical mastery, stubborn persistence, and software taste—suggesting that someone who revolutionized modern computing must necessarily be driven by grand visions of the future. Torvalds immediately rejects the label of "visionary," confessing that he felt slightly out of place during his two days at the TED Conference amid so much sweeping futurism. He makes it clear that he operates without a five-year plan, defining himself strictly as a pragmatic engineer grounded in immediate problem-solving rather than skyward ambition: “I'm perfectly happy with all the people who are walking around and just staring at the clouds and looking at the stars and saying, 'I want to go there.' But I'm looking at the ground, and I want to fix the pothole that's right in front of me before I fall in.”

To illustrate his operational mindset, Torvalds references the classic technology dichotomy between Nikola Tesla and Thomas Edison. While modern culture idolizes Tesla as a romantic, visionary savant—even naming prominent tech companies in his honor—Torvalds places himself squarely in the often-maligned camp of Thomas Edison. He cites Edison’s maxim that genius is 1% inspiration and 99% perspiration, pointing out that despite Edison’s unglamorous, pedestrian reputation and perceived lack of intellectual mysticism, it was Edison's relentless practical labor that fundamentally changed the physical world. Anderson playfully dubs Torvalds the antidote to the conference's theme of bold, audacious dreams, to which Torvalds agrees he is deliberately dialing the rhetoric down.

- Torvalds explicitly disclaims having strategic roadmaps, visionary master plans, or long-range predictions for technology. - Modern cultural sentiment strongly favors Tesla's eccentric genius, yet Torvalds prioritizes tangible real-world transformation over theoretical imagination. - Torvalds embraces Edison's perspiration-heavy work ethic as the truer model for enduring technological progress.

[18:51](https://www.youtube.com/watch?v=o8NPllzkFhE&t=1131s) Commercial Billions, Open Source Limits, and Future Frontiers

Anderson asks Torvalds whether he feels any resentment toward massive corporate entities like Google that have generated billions of dollars in revenue from software he built and gave away for free. Torvalds answers without hesitation that he feels no anger whatsoever. First, he notes that he is personally doing very well financially; second, and more importantly, he recognizes that without embracing the open source ethos and relinquishing control, Linux would never have achieved its global scale or ubiquity. While this path thrust him into uncomfortable obligations such as public speaking, he reflects that the cumulative journey has made him a profoundly contented man who feels he made the right life decisions.

When asked whether open source principles can expand beyond software into broader societal systems like open politics, Torvalds expresses measured skepticism. He explains that software is uniquely suited to open collaboration because code is fundamentally binary: it either works or it fails, providing objective criteria to resolve disagreements. In politics and human affairs, by contrast, straightforward black-and-white logic dissolves into shades of gray and subjective hues. Nevertheless, he points to thriving applications of open principles outside traditional software, citing Wikipedia's worldwide impact as well as the open access renaissance in scientific publishing through platforms like arXiv and open-access journals, which counteract expensive closed publishers. Concluding the session, Anderson observes that since Torvalds is an engineer rather than a prophet, it remains up to the global community to build whatever comes next, thanking Torvalds for giving the world Linux, the underlying infrastructure of the Internet, and the foundation of Android devices.

- Torvalds harbors no bitterness toward commercial tech giants monetizing Linux, viewing open access as the essential catalyst that enabled the system's global adoption. - Software benefits from definitive, objective validation criteria: “Code either works or it doesn't, which means that there's less room for arguments.” - Fields like governance lack objective metrics, complicating the direct transfer of open source methodologies to civic systems. - Scientific research is actively reclaiming its open origins through non-commercial pre-print servers like arXiv and open-access academic publications.

## Notable quotes > [18:51](https://www.youtube.com/watch?v=o8NPllzkFhE&t=1131s) Code either works or it doesn't, which means that there's less room for arguments. — Linus Torvalds

> [14:18](https://www.youtube.com/watch?v=o8NPllzkFhE&t=858s) Good taste is about really seeing the big patterns and kind of instinctively knowing what's the right way to do things. — Linus Torvalds

## Mentioned Android, arXiv, Chris Anderson, CVS, Git, Google, Linux, Nikola Tesla, TED, Thomas Edison, University of Helsinki, Wikipedia

## Video description Linus Torvalds transformed technology twice — first with the Linux kernel, which helps power the Internet, and again with Git, the source code management system used by developers worldwide. In a rare interview with TED Curator Chris Anderson, Torvalds discusses with remarkable openness the personality traits that prompted his unique philosophy of work, engineering and life. "I am not a visionary, I'm an engineer," Torvalds says. "I'm perfectly happy with all the people who are walking around and just staring at the clouds ... but I'm looking at the ground, and I want to fix the pothole that's right in front of me before I fall in."

TEDTalks is a daily video podcast of the best talks and performances from the TED Conference, where the world's leading thinkers and doers give the talk of their lives in 18 minutes (or less). Look for talks on Technology, Entertainment and Design -- plus science, business, global issues, the arts and much more. Find closed captions and translated subtitles in many languages at http://www.ted.com/translate

Follow TED news on Twitter: http://www.twitter.com/tednews Like TED on Facebook: https://www.facebook.com/TED

Subscribe to our channel: http://www.youtube.com/user/TEDtalksDirector

[Source](https://www.youtube.com/watch?v=o8NPllzkFhE)

中文简报

托瓦兹拒绝被贴上“远见者”标签，称Linux与Git源于执着的苦干林纳斯·托瓦兹认为，颠覆性技术的诞生并非来自宏大的远景规划，而是源于执着的工程师默默化解眼前实际的瓶颈。林纳斯·托瓦兹回顾了他当年穿着睡袍写出的单人业余爱好项目，如何演变成现代互联网和超过15亿台安卓设备的基础操作系统。他解释了社区协作中遇到的瓶颈如何促使他发明了Git，以及开源模式为何能够让性格迥异的人在无需彼此喜爱的情况下共同构建成果并取得成功。托瓦兹通过展示优雅的指针设计如何消除基础代码中棘手的极端边缘情况，阐释了他在软件架构中“好品味”的概念。最终，他拒绝接受“远见未来学家”的标签，转而将自己归类于托马斯·爱迪生那种不求光鲜、执着解决眼前实际问题的坚持不懈。林纳斯·托瓦兹创建了如今支撑主要互联网基础设施及超15亿台安卓设备的Linux和Git，其出发点主要是个人需求以及对独立解决问题的偏好，而非出于宏大的战略路线图。 (0:13) Git专门为了打破当时由托瓦兹每两到三个月处理约1000名开发者的内核贡献所带来的人工瓶颈而设计。 (7:13) 托瓦兹将软件工程中的“好品味”定义为一种直观的架构能力，即能够识别系统性模式并从根本上彻底消除边缘情况，而不是仅仅依靠条件分支来进行处理。 (14:18) 托瓦兹反对科技行业对尼古拉·特斯拉这类远见型旷世奇才的浪漫化崇拜，而是将自己务实、立足实际的工程哲学归同于托马斯·爱迪生对汗水与迭代执行的重视。 (16:24)

Source quotes

[18:51] Linus Torvalds: Code either works or it doesn't, which means that there's less room for arguments. [14:18] Linus Torvalds: Good taste is about really seeing the big patterns and kind of instinctively knowing what's the right way to do things.

Original transcript

[0:13] Chris Anderson: This is such a strange thing. [0:15] Your software, Linux, is in millions of computers, [0:18] it probably powers much of the Internet. [0:22] And I think that there are, like, [0:24] a billion and a half active Android devices out there. [0:27] Your software is in every single one of them. [0:30] It's kind of amazing. [0:31] You must have some amazing software headquarters driving all this. [0:37] That's what I thought -- and I was shocked when I saw a picture of it. [0:40] I mean, this is -- [0:41] this is the Linux world headquarters. [0:43] (Laughter) [0:45] (Applause) [0:49] Linus Torvalds: It really doesn't look like much. [0:52] And I have to say, [0:53] the most interesting part in this picture, [0:57] that people mostly react to, [0:59] is the walking desk. [1:00] It is the most interesting part in my office [1:03] and I'm not actually using it anymore. [1:05] And I think the two things are related. [1:09] The way I work is ... [1:13] I want to not have external stimulation. [1:18] You can kind of see, on the walls are this light green. [1:23] I'm told that at mental institutions they use that on the walls. [1:27] (Laughter) [1:29] It's like a calming color, [1:30] it's not something that really stimulates you. [1:35] What you can't see is the computer here, you only see the screen, [1:39] but the main thing I worry about in my computer is -- [1:42] it doesn't have to be big and powerful, although I like that -- [1:46] it really has to be completely silent. [1:49] I know people who work for Google [1:52] and they have their own small data center at home, [1:55] and I don't do that. [1:56] My office is the most boring office you'll ever see. [2:00] And I sit there alone in the quiet. [2:04] If the cat comes up, [2:06] it sits in my lap. [2:08] And I want to hear the cat purring, [2:10] not the sound of the fans in the computer. [2:13] CA: So this is astonishing, [2:14] because working this way, [2:17] you're able to run this vast technology empire -- [2:20] it is an empire -- [2:21] so that's an amazing testament to the power of open source. [2:25] Tell us how you got to understand open source [2:29] and how it lead to the development of Linux. [2:33] LT: I mean, I still work alone. [2:35] Really -- I work alone in my house, [2:38] often in my bathrobe. [2:40] When a photographer shows up, I dress up, [2:42] so I have clothes on. [2:43] (Laughter) [2:45] And that's how I've always worked. [2:47] I mean, this was how I started Linux, too. [2:49] I did not start Linux as a collaborative project. [2:53] I started it as one in a series of many projects [2:57] I had done at the time for myself, [3:01] partly because I needed the end result, [3:03] but even more because I just enjoyed programming. [3:06] So it was about the end of the journey, [3:10] which, 25 years later, we still have not reached. [3:13] But it was really about the fact that I was looking for a project on my own [3:17] and there was no open source, really, on my radar at all. [3:21] And what happened is ... [3:24] the project grows and becomes something you want to show off to people. [3:30] Really, this is more of a, "Wow, look at what I did!" [3:34] And trust me -- it was not that great back then. [3:38] I made it publicly available, [3:40] and it wasn't even open source at that point. [3:42] At that point it was source that was open, but there was no intention [3:47] behind using the kind of open-source methodology that we think of today [3:52] to improve it. [3:53] It was more like, [3:54] "Look, I've been working on this for half a year, [3:58] I'd love to have comments." [4:00] And other people approached me. [4:03] At the University of Helsinki, [4:04] I had a friend who was one of the open source -- [4:08] it was called mainly "free software" back then -- [4:12] and he actually introduced me to the notion that, hey, [4:16] you can use these open-source licenses that had been around. [4:24] And I thought about it for a while. [4:26] I was actually worried about the whole commercial interests coming in. [4:29] I mean, that's one of the worries I think most people who start out have, [4:34] is that they worry about somebody taking advantage of their work, right? [4:39] And I decided, "What the hell?" [4:43] And -- [4:44] CA: And then at some point, [4:46] someone contributed some code that you thought, [4:48] "Wow, that really is interesting, I would not have thought of that. [4:51] This could actually improve this." [4:53] LT: It didn't even start by people contributing code, [4:56] it was more that people started contributing ideas. [4:59] And just the fact that somebody else takes a look at your project -- [5:03] and I'm sure it's true of other things, too, [5:05] but it's definitely true in code -- [5:07] is that somebody else takes an interest in your code, [5:10] looks at it enough to actually give you feedback [5:13] and give you ideas. [5:15] That was a huge thing for me. [5:17] I was 21 at the time, so I was young, [5:20] but I had already programmed for half my life, basically. [5:24] And every project before that had been completely personal [5:28] and it was a revelation when people just started commenting, [5:31] started giving feedback on your code. [5:34] And even before they started giving code back, [5:38] that was, I think, one of the big moments where I said, [5:41] "I love other people!" [5:43] Don't get me wrong -- [5:44] I'm actually not a people person. [5:46] (Laughter) [5:49] I don't really love other people -- [5:52] (Laughter) [5:53] But I love computers, [5:54] I love interacting with other people on email, [5:57] because it kind of gives you that buffer. [6:00] But I do love other people who comment and get involved in my project. [6:06] And it made it so much more. [6:09] CA: So was there a moment when you saw what was being built [6:12] and it suddenly started taking off, [6:14] and you thought, "Wait a sec, this actually could be something huge, [6:17] not just a personal project that I'm getting nice feedback on, [6:20] but a kind of explosive development in the whole technology world"? [6:25] LT: Not really. [6:27] I mean, the big point for me, really, was not when it was becoming huge, [6:30] it was when it was becoming little. [6:33] The big point for me was not being alone [6:37] and having 10, maybe 100 people being involved -- [6:40] that was a big point. [6:42] Then everything else was very gradual. [6:45] Going from 100 people to a million people is not a big deal -- to me. [6:49] Well, I mean, maybe it is if you're -- [6:51] (Laughter) [6:52] If you want to sell your result then it's a huge deal -- [6:55] don't get me wrong. [6:56] But if you're interested in the technology [6:58] and you're interested in the project, [7:00] the big part was getting the community. [7:02] Then the community grew gradually. [7:04] And there's actually not a single point where I went like, [7:08] "Wow, that just took off!" because it -- [7:10] I mean -- it took a long time, relatively. [7:13] CA: So all the technologists that I talk to really credit you [7:16] with massively changing their work. [7:19] And it's not just Linux, [7:21] it's this thing called Git, [7:22] which is this management system for software development. [7:26] Tell us briefly about that and your role in that. [7:30] LT: So one of the issues we had, [7:32] and this took a while to start to appear, [7:35] is when you ... [7:37] When you grow from having 10 people or 100 people working on a project [7:43] to having 10,000 people, which -- [7:45] I mean, right now we're in the situation where just on the kernel, [7:48] we have 1,000 people involved in every single release [7:52] and that's every two months, roughly two or three months. [7:56] Some of those people don't do a lot. [7:58] There's a lot of people who make small, small changes. [8:01] But to maintain this, [8:03] the scale changes how you have to maintain it. [8:06] And we went through a lot of pain. [8:10] And there are whole projects that do only source-code maintenance. [8:17] CVS is the one that used to be the most commonly used, [8:20] and I hated CVS with a passion and refused to touch it [8:24] and tried something else that was radical and interesting [8:27] and everybody else hated. [8:31] CA: (Laughs) [8:32] LT: And we were in this bad spot, [8:34] where we had thousands of people who wanted to participate, [8:37] but in many ways, I was the kind of break point, [8:42] where I could not scale to the point [8:44] where I could work with thousands of people. [8:47] So Git is my second big project, [8:50] which was only created for me to maintain my first big project. [8:55] And this is literally how I work. [8:57] I don't code for -- [9:00] well, I do code for fun -- [9:01] but I want to code for something meaningful [9:04] so every single project I've ever done has been something I needed [9:08] and -- [9:09] CA: So really, both Linux and Git kind of arose [9:13] almost as an unintended consequence [9:15] of your desire not to have to work with too many people. [9:18] LT: Absolutely. Yes. [9:19] (Laughter) [9:20] CA: That's amazing. LT: Yeah. [9:22] (Applause) [9:23] And yet, you're the man who's transformed technology [9:27] not just once but twice, [9:28] and we have to try and understand why it is. [9:31] You've given us some clues, but ... [9:32] Here's a picture of you as a kid, with a Rubik's Cube. [9:37] You mentioned that you've been programming since you were like 10 or 11, [9:41] half your life. [9:43] Were you this sort of computer genius, you know, übernerd, [9:46] were you the star at school who could do everything? [9:49] What were you like as a kid? [9:52] LT: Yeah, I think I was the prototypical nerd. [9:54] I mean, I was ... [9:57] I was not a people person back then. [9:59] That's my younger brother. [10:01] I was clearly more interested in the Rubik's Cube [10:04] than my younger brother. [10:06] (Laughter) [10:07] My younger sister, who's not in the picture, [10:10] when we had family meetings -- [10:12] and it's not a huge family, but I have, like, a couple of cousins -- [10:16] she would prep me beforehand. [10:19] Like, before I stepped into the room she would say, [10:22] "OK. That's so-and-so ..." [10:25] Because I was not -- [10:28] I was a geek. [10:29] I was into computers, [10:30] I was into math, [10:31] I was into physics. [10:33] I was good at that. [10:34] I don't think I was particularly exceptional. [10:37] Apparently, my sister said [10:39] that my biggest exceptional quality was that I would not let go. [10:47] CA: OK, so let's go there, because that's interesting. [10:50] You would not let go. [10:51] So that's not about being a geek and being smart, [10:54] that's about being ... stubborn? [10:57] LT: That's about being stubborn. [10:58] That's about, like, [11:01] just starting something [11:03] and not saying, "OK, I'm done, let's do something else -- [11:08] Look: shiny!" [11:10] And I notice that in many other parts in my life, too. [11:15] I lived in Silicon Valley for seven years. [11:18] And I worked for the same company, in Silicon Valley, [11:22] for the whole time. [11:24] That is unheard of. [11:27] That's not how Silicon Valley works. [11:29] The whole point of Silicon Valley is that people jump between jobs [11:32] to kind of mix up the pot. [11:35] And that's not the kind of person I am. [11:37] CA: But during the actual development of Linux itself, [11:39] that stubbornness sometimes brought you in conflict with other people. [11:44] Talk about that a bit. [11:46] Was that essential to sort of maintain the quality of what was being built? [11:51] How would you describe what happened? [11:54] LT: I don't know if it's essential. [11:57] Going back to the "I'm not a people person," -- [12:00] sometimes I'm also ... [12:04] shall we say, [12:05] "myopic" when it comes to other people's feelings, [12:09] and that sometimes makes you say things that hurt other people. [12:14] And I'm not proud of that. [12:18] (Applause) [12:19] But, at the same time, it's -- [12:22] I get people who tell me that I should be nice. [12:26] And then when I try to explain to them that maybe you're nice, [12:32] maybe you should be more aggressive, [12:34] they see that as me being not nice. [12:38] (Laughter) [12:40] What I'm trying to say is we are different. [12:42] I'm not a people person; [12:44] it's not something I'm particularly proud of, [12:46] but it's part of me. [12:48] And one of the things I really like about open source [12:50] is it really allows different people to work together. [12:56] We don't have to like each other -- [12:58] and sometimes we really don't like each other. [13:00] Really -- I mean, there are very, very heated arguments. [13:04] But you can, actually, you can find things that -- [13:08] you don't even agree to disagree, [13:10] it's just that you're interested in really different things. [13:13] And coming back to the point where I said earlier [13:16] that I was afraid of commercial people taking advantage of your work, [13:20] it turned out, and very quickly turned out, [13:23] that those commercial people were lovely, lovely people. [13:26] And they did all the things that I was not at all interested in doing, [13:30] and they had completely different goals. [13:32] And they used open source in ways that I just did not want to go. [13:37] But because it was open source they could do it, [13:39] and it actually works really beautifully together. [13:42] And I actually think it works the same way. [13:44] You need to have the people-people, the communicators, [13:47] the warm and friendly people [13:49] who like -- [13:50] (Laughter) [13:52] really want to hug you and get you into the community. [13:55] But that's not everybody. [13:56] And that's not me. [13:58] I care about the technology. [13:59] There are people who care about the UI. [14:01] I can't do UI to save my life. [14:04] I mean, if I was stranded on an island [14:07] and the only way to get off that island was the make a pretty UI, [14:11] I'd die there. [14:12] (Laughter) [14:13] So there's different kinds of people, [14:15] and I'm not making excuses, I'm trying to explain. [14:18] CA: Now, when we talked last week, [14:20] you talked about some other trait that you have, [14:22] which I found really interesting. [14:24] It's this idea called taste. [14:25] And I've just got a couple of images here. [14:28] I think this is an example of not particularly good taste in code, [14:32] and this one is better taste, [14:35] which one can immediately see. [14:37] What is the difference between these two? [14:40] LT: So this is -- [14:41] How many people here actually have coded? [14:45] CA: Oh my goodness. [14:47] LT: So I guarantee you, [14:48] everybody who raised their hand, [14:49] they have done what's called a singly-linked list. [14:52] And it's taught -- [14:55] This, the first not very good taste approach, [14:58] is basically how it's taught to be done when you start out coding. [15:02] And you don't have to understand the code. [15:04] The most interesting part to me [15:06] is the last if statement. [15:10] Because what happens in a singly-linked list -- [15:13] this is trying to remove an existing entry from a list -- [15:16] and there's a difference between if it's the first entry [15:20] or whether it's an entry in the middle. [15:22] Because if it's the first entry, [15:24] you have to change the pointer to the first entry. [15:27] If it's in the middle, [15:28] you have to change the pointer of a previous entry. [15:31] So they're two completely different cases. [15:33] CA: And that's better. [15:34] LT: And this is better. [15:35] It does not have the if statement. [15:38] And it doesn't really matter -- [15:40] I don't want you understand why it doesn't have the if statement, [15:44] but I want you to understand [15:45] that sometimes you can see a problem in a different way [15:48] and rewrite it so that a special case goes away [15:51] and becomes the normal case. [15:53] And that's good code. [15:55] But this is simple code. [15:56] This is CS 101. [15:58] This is not important -- although, details are important. [16:01] To me, the sign of people I really want to work with [16:05] is that they have good taste, which is how ... [16:09] I sent you this stupid example [16:10] that is not relevant because it's too small. [16:13] Good taste is much bigger than this. [16:15] Good taste is about really seeing the big patterns [16:19] and kind of instinctively knowing what's the right way to do things. [16:24] CA: OK, so we're putting the pieces together here now. [16:27] You have taste, [16:30] in a way that's meaningful to software people. [16:32] You're -- [16:33] (Laughter) [16:35] LT: I think it was meaningful to some people here. [16:39] CA: You're a very smart computer coder, [16:42] and you're hellish stubborn. [16:45] But there must be something else. [16:47] I mean, you've changed the future. [16:48] You must have the ability of these grand visions of the future. [16:51] You're a visionary, right? [16:53] LT: I've actually felt slightly uncomfortable at TED [16:55] for the last two days, [16:57] because there's a lot of vision going on, right? [17:00] And I am not a visionary. [17:02] I do not have a five-year plan. [17:04] I'm an engineer. [17:05] And I think it's really -- [17:07] I mean -- I'm perfectly happy with all the people [17:09] who are walking around and just staring at the clouds [17:12] and looking at the stars and saying, "I want to go there." [17:15] But I'm looking at the ground, [17:16] and I want to fix the pothole that's right in front of me [17:19] before I fall in. [17:20] This is the kind of person I am. [17:22] (Cheers) [17:23] (Applause) [17:25] CA: So you spoke to me last week about these two guys. [17:29] Who are they and how do you relate to them? [17:32] LT: Well, so this is kind of cliché in technology, [17:35] the whole Tesla versus Edison, [17:37] where Tesla is seen as the visionary scientist and crazy idea man. [17:42] And people love Tesla. [17:45] I mean, there are people who name their companies after him. [17:48] (Laughter) [17:51] The other person there is Edison, [17:54] who is actually often vilified for being kind of pedestrian [17:57] and is -- [17:59] I mean, his most famous quote is, [18:02] "Genius is one percent inspiration and 99 percent perspiration." [18:07] And I'm in the Edison camp, [18:08] even if people don't always like him. [18:11] Because if you actually compare the two, [18:14] Tesla has kind of this mind grab these days, [18:18] but who actually changed the world? [18:21] Edison may not have been a nice person, [18:25] he did a lot of things -- [18:28] he was maybe not so intellectual, [18:30] not so visionary. [18:33] But I think I'm more of an Edison than a Tesla. [18:37] CA: So our theme at TED this week is dreams -- [18:40] big, bold, audacious dreams. [18:41] You're really the antidote to that. [18:43] LT: I'm trying to dial it down a bit, yes. [18:45] CA: That's good. [18:46] (Laughter) [18:47] We embrace you, we embrace you. [18:51] Companies like Google and many others have made, arguably, [18:53] like, billions of dollars out of your software. [18:56] Does that piss you off? [18:57] LT: No. [18:58] No, it doesn't piss me off for several reasons. [19:01] And one of them is, I'm doing fine. [19:03] I'm really doing fine. [19:04] But the other reason is -- [19:06] I mean, without doing the whole open source and really letting go thing, [19:11] Linux would never have been what it is. [19:14] And it's brought experiences I don't really enjoy, public talking, [19:19] but at the same time, this is an experience. [19:21] Trust me. [19:23] So there's a lot of things going on that make me a very happy man [19:28] and thinking I did the right choices. [19:30] CA: Is the open source idea -- [19:32] this is, I think we'll end here -- [19:34] is the open source idea fully realized now in the world, [19:38] or is there more that it could go, [19:41] are there more things that it could do? [19:44] LT: So, I'm of two minds there. [19:47] I think one reason open source works so well in code [19:51] is that at the end of the day, [19:53] code tends to be somewhat black and white. [19:57] There's often a fairly good way to decide, [20:01] this is done correctly and this is not done well. [20:05] Code either works or it doesn't, [20:07] which means that there's less room for arguments. [20:12] And we have arguments despite this, right? [20:16] In many other areas -- [20:17] I mean, people have talked about open politics and things like that -- [20:21] and it's really hard sometimes to say [20:24] that, yes, you can apply the same principles in some other areas [20:28] just because the black and white turns into not just gray, [20:33] but different colors. [20:36] So, obviously open source in science is making a comeback. [20:40] Science was there first. [20:42] But then science ended up being pretty closed, [20:45] with very expensive journals and some of that going on. [20:48] And open source is making a comeback in science, [20:52] with things like arXiv and open journals. [20:59] Wikipedia changed the world, too. [21:01] So there are other examples, [21:02] I'm sure there are more to come. [21:05] CA: But you're not a visionary, [21:07] and so it's not up to you to name them. [21:09] LT: No. [21:10] (Laughter) [21:11] It's up to you guys to make them, right? [21:13] CA: Exactly. [21:14] Linus Torvalds, [21:15] thank you for Linux, thank you for the Internet, [21:18] thank you for all those Android phones. [21:20] Thank you for coming here to TED and revealing so much of yourself. [21:23] LT: Thank you. [21:24] (Applause)