Linus 訪談,在 Linux 誕生 30 週年之際

Linux 超級開源項目 30 週年 - Jim Zemlin 獨家專訪 Linus Torvalds

官方網站,https://gotc.oschina.net/

今年是 Linux 誕生 30 週年,Linux 從一個開源操作系統項目,擴展成爲今日無數個不同的開源項目,從超級電腦到雲計算平臺到智能手機,Linux 影響力可以說無處不在,今天的企業比以往任何時候都更依賴通過開源爲客戶提供價值!在本次由 Linux 基金會執行董事 Jim Zemlin 主持的獨家訪談中,Linux 之父 Linus Torvalds 將會回顧這個旅程中的難忘經歷,並且分享 Linux 面臨的挑戰和未來的展望。同時,Linus 還會跟中國開發者和企業分享寶貴的開源經驗和建議。

2021 年 7 月 9 日

內容摘要

Linus Torvalds 在與 Jim 的訪談中講述了自己 30 年來,在維護 Linux 內核過程中遇到的困難和挑戰,並對想要參與 Linux 內核貢獻的中國開發者提出了自己的建議。Linus Torvalds 表示,Linux 內核發展最大的挑戰是適配日新月異的計算機硬件,同樣,所有困難且前沿的創新技術也大多集中在硬件上。他認爲,文化和語言的壁壘是中國以及亞洲開發者遇到的挑戰,希望中國開發者學習英語。中國擁有自己的 CPU 指令集架構,開發者還要更多的去了解 CPU 等硬件知識。他還希望想要參與 Linux 內核或其他任何開源項目貢獻的開發者,可以一步一個腳印,從小事做起,從個人感興趣的事情開始做,慢慢找到自己能夠爲項目做出真正貢獻的部分。

採訪的全文

Hello everybody, my name is jim zemlin. I'm the executive director of the Linux foundation and i'm joined today by the creator of linux linus torvalds. The linux foundation is really glad to be participating in today's event, and today is really a special day because we're celebrating this year thirty years of linux. And what amazing journey has been. Linux is the most successful open source project in the history of computing, and it's the most successful software in the history of computing. Not only has linux enabled amazing innovation, but it really ushered in open source as a mainstream way that any technology product or service gets. China is an important part that that open source ecosystem as well. And that's why we're so delighted to be here today. I want to start off my conversation with linus by asking him the most obvious question on the 30th anniversary of the project. Did you think that it would really change the entire tech industry this much when you first got started.

大家好,我叫吉姆澤林。我是 Linux 基金會的執行董事,今天 linux 的創造者 linus torvalds 加入了我的行列。Linux基金會真的很高興參加今天的活動,今天真的是一個特殊的日子,因爲我們正在慶祝今年的linux三十週年。多麼令人驚歎的旅程。Linux是計算史上最成功的開源項目,也是計算史上最成功的軟件。linux不僅實現了驚人的創新,還真正將開源作爲任何技術產品或服務的主流方式。中國也是開源生態系統的重要組成部分。正因如此,我們今天很高興來到這裏。在項目30週年之際,我想通過問他一個最明顯的問題來開始我和萊納斯的對話。當你剛開始的時候。你認爲這真的會改變整個科技行業嗎?

Hi Jim, Obviously no. When i started thirty years ago. Linux was really just a personal project for my own enjoyment. and to learn how the pc i had just bought for it. There were no big plans. There wasn't even big expectation that a lot of people would start joining me. when i made it available for the first time it was literally more of a hey look at what i've done. This is interesting. Maybe somebody else might be interested in it too and it took a while. I mean, it took a while before people really started participating but It actually was what made the project even more interesting for me. It went from something that was just my personal private project to something where it was not just about the technology. It was about communication, It was about working together. and it really took off, thanks to making it public.

嗨,吉姆,顯然不是。當我三十年前開始的時候。Linux實際上只是一個我自己享受的個人項目。用來了解我剛買的電腦。沒有什麼大計劃。甚至沒有非常期望很多人會開始加入我。當我第一次提供它時,它實際上更像是對我所做的事情的一種審視。這很有趣。也許其他人也對它感興趣,而且花了一段時間。我的意思是,人們花了一段時間才真正開始參與,但這實際上讓這個項目對我來說更加有趣。它從一個只是我個人的私人項目變成了一個不僅僅是關於技術的項目。它是關於溝通的,它是關於合作。多虧了公開,它才真正起飛。

Yeah, i mean one of the cool things about linux or any open source project is, anybody from anywhere no matter where you live or work for, It can participate in the project. Anybody for whatever reason they want can go use the software to create whatever it is they think is interesting. And certainly that's been a big part of the success. When you think about the future. Linux is now in automobiles and in embedded systems and all sorts of places, what do you think is in sorts for the future of life.

是的,我的意思是關於linux或者任何開源項目的一個很酷的事情是,來自任何地方的任何人,不管你在哪裏生活或者工作,它都可以參與這個項目。任何人,不管出於什麼原因,都可以使用這個軟件去創造任何他們認爲有趣的東西。這當然是成功的一大部分。當你想到未來。Linux現在出現在汽車、嵌入式系統和各種各樣的地方,你認爲未來的生活是什麼樣的。

So that's not how i actually work. For me, the future is literally about two releases away, look about four to six months ahead and look at it. What are the immediate problems and We want to solve, because quite a often what happens is that, the future of the operating system is not so much determined by me or even by the other developers. It's determined by harder improvements, but also by new usage cases. So i don't even want to say where we're do. I want to be in another thirty years, because I hoped that the use cases and the hardware will all be so surprising and so different. Linux will have then adapted to a whole new world, and any plans we do today will not be relevant, and even ten years much less thirty.

其實我實際上不是這樣工作的。對我來說,未來實際上是兩個版本號的距離,看看未來的四到六個月,看看它。有哪些迫在眉睫的問題是我們想要解決的,因爲通常情況下,操作系統的未來既不是由我決定的,也不是由其他開發人員決定的。它取決於努力的改進,也取決於新的應用場景。所以我甚至不想說我們在做什麼。我想再過三十年,因爲我希望場景和硬件都將如此令人驚訝和如此不同。到那時,Linux將會適應一個全新的世界,我們今天所做的任何計劃都是不相關的,甚至十年內就會發生。

I think that's one of the aspects of the project. That's so impressive is that, you focus on creating the most amazing support for sort of whatever comes your way in terms of hardware innovation or new use cases, in that single focus on creating the world's greatest your operating system kernel. I think it is just amazingly Impressive. Speaking of success in how impressive it is. What do you think are some of the challenges on the horizon for linux or for open source in general.

我認爲這是這個項目的一個方面。令人印象深刻的是,你專注於爲硬件創新或新用例方面的任何東西創造最驚人的支持,專注於創造世界上最偉大的操作系統內核。我認爲這令人驚訝地印象深刻。說到成功是多麼令人印象深刻。你認爲linux或開源軟件面臨的挑戰有哪些?

So i think linux in particular one of the big challenges will be, how more slowly fading, and sometimes not so slowly, which means that we have gotten used to over the past thirty years. We've gotten used to the fact that, hardware has been just growing enormously. And that's not necessarily as true anymore. And i don't mean that, I mean obviously hardware in some sense has been shrinking. with cell phones, for example being a big issue for linux these days, but the capabilities has still be growing. And i think that's just some degree changing. And i think it's going to be a farely different landscape, when one people really don't expect hardware should get more capabilities every year, and it's going to change how suffered development is being done. And i think linux in particular because we've been reacting so much to hardware changes. This is going to probably be the biggest change we have coming up, Is how the hardware landscape development changes.  because of this. I don't know how much it's going to affect other areas.

所以我認爲特別是linux,最大的挑戰之一是,如何更慢地衰落,有時不那麼慢,這意味着我們已經習慣了過去三十年。我們已經習慣了這樣一個事實,硬件正在飛速發展。現在不一定了。我不是這個意思,我是說很明顯硬件在某種意義上一直在萎縮。例如,手機現在是linux的一個大問題,但是它的功能仍然在增長。我認爲這只是某種程度上的改變。我認爲這將是一個完全不同的局面,當一個人真的不希望硬件每年都有更多的功能時,這將改變開發的方式。我認爲特別是linux,因爲我們對硬件的變化反應如此之大。這可能是我們接下來最大的變化,就是硬件環境的發展變化。正因爲如此。不知道會對其他領域造成多大影響。

I think it's amazing to see just as semiconductor technology is shifting expanding. All sorts of new silicon architectures are coming and going. How consistent linux has been in terms of working within each of those ecosystems like any. Any secret to how you're able to go from intel architecture to RISC-V architecture to ARM architecture. Like what.What's the secret.

我認爲,看到半導體技術正在轉移和擴張,這是令人驚訝的。各種新的硅架構層出不窮。就像任何生態系統一樣,linux在每一個生態系統中的工作是多麼的一致。從英特爾架構到RISC-V架構再到ARM架構,有什麼祕訣嗎?比如什麼?有什麼祕密?

I think one of the big issues has been that, for me all the changes and a lot of kernel developers too, all the different architectures all that huge changes in SMP's capabilities have been seen not as like challenges to be overcome, but they have been what makes doing kernel so interesting. The fact that, we have something like almost thirty different architectures that we support has been part of the fun of doing linux. And that's actually one of the worries i have, if hardware development slows down. We want to have as many new exciting areas to work with as we traditionally have. I actually really i think having multiple architectures has not only been an interesting challenge for us. It's also being what has made linux very solid. Because in trying to support not just one architecture or we have had to basically rearchitects our own internal code so that it is more flexible.And it has obviously been a lot of work, but it's been what has made linux so successfully in the end.

我認爲最大的問題之一是,對我和許多內核開發人員來說,所有不同的體系結構SMP(對稱多核處理)能力的巨大變化都被認爲不是需要克服的挑戰,但它們是讓做內核變得如此有趣的原因。事實上,我們幾乎有30種不同的架構支持,這也是做linux的樂趣之一。如果硬件開發速度變慢,這實際上是我擔心的問題之一。我們希望有和傳統一樣多的令人興奮的新領域可以合作。我真的認爲擁有多種架構不僅對我們來說是一個有趣的挑戰。這也是linux變得非常穩固的原因。因爲在試圖支持不僅僅是一個架構時,或者我們必須重新構建我們自己的內部代碼,以便它更加靈活。很明顯,這是一項很大的工作,但正是它最後讓linux如此成功。

Yeah, I think it's interesting, you know The multi architecture support creates this positive feedback loop where semiconductor manufacturers and people creating new silicon can sort of get to value faster. Because linux is so stable and architecture to support multiple architectures which begins more interesting hardware because they can get to market faster and so on and so forth. So i think you'll be having access to a lot of interesting new hardware for a long time. I want to say a quick question about china, you know more chinese developers want to join the kernel project. you certainly  know open source is welcome anyone. To anyone,what are the skills and for attitude that developers should have, when participating in a project like linux, what are sort of the do's and delts.

是的,我認爲這很有趣,你知道,多架構支持創造了這種正反饋循環,半導體制造商和創造新硅的人可以更快地實現價值。因爲linux非常穩定,而且架構支持多種架構,這就開始了更有趣的硬件,因爲它們可以更快地上市等等。所以我認爲你將在很長一段時間內接觸到很多有趣的新硬件。我想簡單問一個關於中國的問題,你知道更多的中國開發者想加入內核項目。你當然知道開源是歡迎任何人的。對任何人來說,當參與像linux這樣的項目時,開發人員應該具備什麼樣的技能和態度,什麼樣的做法和模式。

Well, we actually have tried to avoid having any very strict rules about what you should do and what you shouldn't do. and different regions have very different problems. I mean one of the issues obviously with china and Asia in general has been that, linux has traditionally been developed. Using email and speaking english. So when you have any new region that gets involved, one of the big issues ends up being just communication. And that doesn't mean and i'm not trying to say at all, that now everybody should try to learn english in order to participate in the kernel community, but it does mean for example, that one of the things we look for is kind of translator, a person who is good at communicating and can communicate with both sides, and acts as kind of a go between the kind of traditional linux kernel community .and i'm sure there is a very vibrant kernel community inside china that we just don't see. Because of the language barriers. so that's like one of the things that is very specific to china will not specific to chinese itself, but is one of the main issues for countries like china where there are cultural language barriers. The other thing that obviously you need to have, is just the deep technical knowledge about cpu architectures. Because kernels tend to be kind of unusual, in the sense that they are more about the hardware than a lot of other software projects. And i think this is an area that china certainty has a lot of depth with their own cpu architecture is going on. But it's still, If you want to be a kernel developer, it is something you have to be very aware of that. And you have to be interested in hardware, not necessary. Just the CPU. About half of the kernel is actually drivers for The other hardware around the CPU.

嗯,我們實際上已經試圖避免對你應該做什麼和不應該做什麼有任何非常嚴格的規定。不同的地區有非常不同的問題。我的意思是,中國和亞洲的一個明顯問題是,linux傳統上是被開發出來的。使用電子郵件和說英語。因此,當你有任何新的領域參與進來時,最大的問題之一就是溝通。這並不意味着,我也不是想說,現在每個人都應該努力學習英語,以便參與內核社區,但這確實意味着,例如,我們尋找的東西之一是一種翻譯器,一個善於溝通並能與雙方溝通的人,並充當傳統linux內核社區的中間人。我相信在中國有一個非常活躍的內核社區,這是我們看不到的。因爲語言障礙。因此,這就像是中國特有的事情之一,而不是中國本身特有的,而是像中國這樣存在文化語言障礙的國家的主要問題之一。另一件你顯然需要具備的事情,就是關於cpu架構的深厚技術知識。因爲內核往往有點不尋常,從某種意義上說,內核更多的是關於硬件,而不是其他軟件項目。我認爲這是一個領域,中國肯定有很大的深度與他們自己的cpu架構正在進行。但它仍然是,如果你想成爲一名內核開發人員,你必須非常清楚這一點。你必須對硬件感興趣,這不是必須的。只有CPU。大約一半的內核實際上是中央處理器周圍其他硬件的驅動程序。

Yeah, i mean it's a big commitment to sort of learn. A difficult technology focused on hardware. And certainly it seems intimidating to get started. I mean what words of encouragement can you offer to all the chinese developers are out there current and future ones, about how to participate in either linux or open source in general.

是的,我是說這是一個很大的學習承諾。專注於硬件的高難度技術。當然,開始似乎有點嚇人。我的意思是,對於當前和未來的中國開發者,關於如何參與linux或開源,你能給他們什麼鼓勵的話。

I think the main issue is don't try to do everything at once and  and start small and start with something you are personally interested in. And i think you have something to really offer, and it doesn't have to be. I mean it just going back to my own starting point. It really does not have to be something where you think that you are going to change the world. Make it be something that you are passionate about that you really think you can bring something to the table. And even if it seems small and and not very important in the big picture, that's how you get started. And that's how you get into any project whether it's the kernel or other open source project. Find the wish, that you care about and that you want to really improve. That's how you learn how the processors work, but it's also how you make other people aware of your skill sets and that's how you grow up person.

我認爲主要的問題是不要試圖一次做所有的事情,從小事做起,從你個人感興趣的事情開始。我認爲你真的可以提供一些東西,但也不一定。我的意思是它只是回到我自己的起點。它真的不一定是你認爲你會改變世界的東西。讓它成爲你充滿熱情的東西,你真的認爲你可以帶來一些東西。即使它看起來很小,在大局中也不太重要,這就是你如何開始的。這就是你進入任何項目的方式,無論是內核還是其他開源項目。找到你關心的,你想真正提高的願望。這是你學習處理器如何工作的方式,但也是你如何讓別人意識到你的技能組合,這就是你如何成長的方式。

I agree, and that's terrific advice. Well, I want to thank you for joining us here today. Linus, you know it's thirty years in, it's hard for me to even imagine, you've been doing the same job continuously for thirty years. You pause a little bit to write Git, which is impressive in and of itself, but to have the fortitude to, every three months release the linux kernel. For this one is pretty amazing. and it's just an honor always to work with you, and to all you chinese developers out there. You're looking at someone, who proof that, the world is non zero sum game, by working together. You can change the world. You're a shining example to all of us, so thank you. Now you're embarrassing me, but thank you.

我同意,這是很好的建議。嗯,我想感謝你今天加入我們。萊納斯,你知道已經三十年了,我甚至很難想象,你已經連續三十年做同樣的工作了。你稍微停下來寫Git,這本身就令人印象深刻,但是要有毅力,每三個月發佈一次linux內核。因爲這一個非常驚人。和你們一起工作是我的榮幸,也是所有中國開發者的榮幸。你在看一個人,他證明,世界是非零和遊戲,靠的是共同努力。你可以改變世界。你是我們所有人的光輝榜樣,所以謝謝你。現在你讓我很尷尬,但是謝謝你。

Thank you.

謝謝你。

發表評論
所有評論
還沒有人評論,想成為第一個評論的人麼? 請在上方評論欄輸入並且點擊發布.
相關文章