如果大學不把基礎研究和應用研究區分開，永遠無法培養出真正需要的人才

文/約翰·霍普克羅夫特（John Hopcroft，美） 譯/李睿 楊韞利

2020年12月19日，圖靈獎獲得者、中國科學院外籍院士、北京大學訪問講席教授、前沿計算研究中心主任約翰·霍普克羅夫特（John Hopcroft）受邀在2020中國計算機教育大會作題為《以教育質量提升助力中國經濟發展》的主旨報告，介紹他與中國工程院院士、北京大學教授高文于2018年共同發起“高校計算機專業優秀教師獎勵計劃”的初衷。以下由北京大學前沿計算研究中心翻譯整理自報告原文，小標題為編者所加，以饗讀者。

今天我們在這里表彰和獎勵一批杰出的教師。54位獲獎教師是通過嚴格的評審程序，從中國的44所優秀大學中選拔出來的。

中國大學需要專注提升教學質量

20年前，中國的父母們開始意識到，他們的孩子需要接受大學教育才能找到一份理想的工作。為了滿足人們對高等教育的迫切需要，中國每年新建近50所大學，增加了70萬名教師來教授新增的6000萬本科生。如今，中國已經具備了為眾多學生提供教育的基本條件，重點開始轉向提升教育的質量。

中國的一些頂尖大學創建了精英班，每個班選拔25—30名學生。然而，這些精英班的規模太小了，遠遠不能滿足需求。另外，許多精英班的學生畢業后選擇去海外深造，留在中國申請理工科博士學位的高質量學生相對有限。

人才在世界各地是均勻分布的，但機會卻并非如此。中國擁有全世界1/5的人才，所有的優秀學生都希望進入頂尖大學，接受最好的教育。我曾作為客座教授，在中國的幾所大學教過剛入學的新生，他們的水平完全不遜于美國大學的新生，甚至更好。然而，當我教大三的學生時，我發現他們在知識水平和能力上已落后于美國同年級的學生。我意識到，這可能是中國大學的教育質量出現了問題。

曾有一所中國C9（九校聯盟）高校邀請我幫他們的計算機系招聘老師。兩年后，這些新老師都離開了這所大學。我詢問原因，得到的答復是，學校的大環境只注重研究經費和論文發表數量，對教學質量的關注度非常低。我和其他一些中國大學的老師們交流過這個現象，他們也有同感。

中國的大學需要提升本科教育質量。大學校長每屆任期5年，他們普遍認為，提高大學的國際排名是校長任期內工作成效的最佳體現之一。然而，現有的國際排名評價體系主要是基于研究經費的多少和發表論文的數量，這給老師們帶來了壓力。為了晉升，他們把大量時間花在經費申請和論文發表上。我想說的是，如果校長們希望提高學校的排名和聲譽，他們需要專注于教學質量的提升。

應運而生的“高校計算機專業優秀教師獎勵計劃”

一些新入職的助理教授要為資深教授工作，而不是指導他們自己的研究生。無論是經費申請還是論文發表都與本科生教育沒有直接關系。很明顯，大學培養下一代人才的使命不知何故被忽視了。

創建“高校計算機專業優秀教師獎勵計劃”就是為了表彰注重教學的優秀教師，并希望大學能夠調整重心，重視本科生教學質量。在李克強總理的鼓勵和教育部的支持下，以評估老師教學質量為目的的“高校計算機專業優秀教師獎勵計劃”應運而生。對大學本科教學質量的評估需要一個衡量標準。該計劃的評估標準基于對老師教學質量的評價，而非研究經費和論文發表數量。評估的過程得到了中國一些單位的捐贈支持，他們將此視為一項長期投資，以滿足對其發展至關重要的人才的需求。2020年的捐贈單位包括：字節跳動、科大訊飛、騰訊基金會、平安科技。

評估涉及兩個問題：授課質量和課程內容。為了評估授課質量，“獎勵計劃”評審專家組分派兩位專家隨機旁聽每位候選人的兩堂課，并使用記分卡為所聽的課程評分。記分卡包括以下內容：教師對講授課程內容和教材的掌握度；教師是否對教學內容保持興奮態度；教師是否和學生有良好的課堂互動；全程認真聽課的學生比例（或有多少學生在低頭玩手機）；學生是否準時到達教室上課；教師是否使用幻燈片或黑板。最后，專家會針對課堂的整體情況提供一段評價。

評估工作開始于2018年。作為試點，我們首先對中國C9高校的計算機專業教學情況進行了評估。2019年，該計劃的評估范圍擴大到國內的44所大學。

我們原計劃在2020年春季表彰這54位獲獎老師，同時啟動新一輪的評估工作，但新冠肺炎疫情打亂了我們的計劃。

霍普克羅夫特教授在2020中國計算機教育大會講話

我們還計劃通過評估結果告訴大學的校長們，他們的學校在計算機專業的本科教育上表現如何：排前1/4，前一半還是后一半。然而，這一年校長們忙于在疫情的侵襲下維持大學的正常運轉，可能沒有多余的精力關注我們想要傳達的信息。

2021年春季開始，我們將重新啟動評估工作，并希望創建一份針對教學質量的大學排名，這樣家長們就會知道孩子申請哪所大學能夠獲得最好的教育。中國大學的校長們有著極高的素質和才能，他們完全有能力迅速將學校的教學質量提升到斯坦福、伯克利和康奈爾等大學的水平。

大學的使命是培養人才

大學的使命是培養下一代人才，而不是搞研究。那么，為什么我們又希望老師們從事研究呢？教師的職業生涯大約是40年，我們希望老師們能夠在其領域保持前沿水平，不希望他們一成不變地教著40年前的教材，沒有任何更新和改變。優秀教師會緊跟其領域的發展，不斷學習，參與研究就是表現之一。這里強調的不是老師的研究水平高低。比如，有的老師可能只是針對一個已知結果尋找更簡單的證明，或對一些經典的理論知識找到一種更好的教學方法。我們看重的是老師是否有求知欲，是否會隨著所在領域的變化而發展，是否關心學生的成功。

第二次世界大戰之后，美國意識到科學在戰爭中起到的重要作用，決定大力發展科學。他們提出了兩類科學研究的劃分：應用科學和基礎科學。

應用科學是為解決具體問題而進行的研究。美國有足夠多的研究員和工程師。因此，當一家企業甚至是整個國家遇到問題，他們可以聘請所需的研究人員來解決。如果企業有需要，他們可以直接聘請人才；如果政府有需要，他們會通過國家實驗室聘請研究人員。

與之不同的是，基礎科學不以具體問題為導向，而單純是好奇心驅動的研究。這部分研究適合在大學開展。同時，大學還將承擔培養下一代研究人員的責任。資助基礎科學研究是極其重要的。它不僅培養了下一代研究人員，還可能在不經意間開創一個全新的領域，新的產業和上百萬的就業機會也會應運而生。

目前，中國還沒有培養出足夠數量的科學和工程類人才來滿足其對于應用科學研究的需要。因此，中國的大學教師必須同時兼顧應用科學和基礎科學的研究。然而，如果大學不把應用科學和基礎科學的研究區分開，他們永遠無法培養出真正需要的人才。大學的經費應當用來資助基礎科學研究，應用科學研究應由有相關需求的企業和政府機構來資助。最終，應用科學研究應該走出校園，就像麻省理工學院創建的林肯實驗室、斯坦福大學創建的斯坦福研究院那樣。

未來，教育是重中之重

未來，教育是重中之重。那些在人工智能等重要領域培養出優秀人才的國家將會吸引企業投資，開創新興產業，推動國家經濟發展。國家的價值將由人才決定，而不是煤炭、石油、木材等物質資源。

人工智能的出現將淘汰大量低水平的工作，從事高水平的工作需要大學教育。當代學生在走出校園的若干年后，可能在工業界或政府擔任重要職位，到那時，狹隘的科學教育帶來的問題將逐漸顯現。正因如此，即使是理工學科，在培養方案中加入歷史、人文以及社會科學課程也是非常重要的。

大學教育不僅是為了讓學生畢業后得到一份工作，也是為了讓他們更好地體驗和享受人生。在職業生涯之余，他們也有機會享受音樂、運動和旅行。他們需要的不僅是理工學科的知識，還有綜合素質的培養。

做一名教師是很有意義的工作，因為40年后你們退休的時候，終有一天你們會對自己幫助過的學生以及為國家作出的貢獻感到驕傲。

中國有一句名言很適合今天的大會：“一年之計，莫如樹谷；十年之計，莫如樹木；終身之計，莫如樹人?！苯裉煸谧膬炐憬處焸冋谧龅木褪恰鞍倌陿淙恕钡墓ぷ鳌獮閲遗囵B優秀人才。他們意識到了教育的重要性，希望為中國的長遠發展貢獻自己的力量。

December 19th, 2020, the Turing Award winner John Hopcroft, a visiting professor at Peking University and the Director of PKU Center on Frontiers of Computing Studies, made a keynote speech on The Role of Improving Education in China’s Economic Growth at the 2020 Computer Education Conference of China. Below is his full thought-provoking speech.

Today we are here to recognize and honor a group of outstanding faculty. They have been selected by a rigorous process where faculty who care about quality of teaching sat in on lectures of computer science faculty at the top 44 universities in China and selected the 54 teachers that we are recognizing here today.

Universities in China need to focus on improving teaching quality

Twenty years ago parents began to realize that their children needed a college education to get a good job. To prevent unhappy citizens, China needed to provide the capacity to meet this demand. China created approximately 50 universities per year and added 700,000 faculty to handle an additional sixty million undergraduates. Now that the educational capacity has been met, the focus has switched to creating quality.

A number of top universities in China created elite programs with twenty-five to thirty students that are world class. However, the elite programs are too small for China’s needs. Many of the graduates from these elite programs go abroad for their PhD. There are not enough high-quality applicants for China’s PhD programs in science and engineering.

Talent is uniformly distributed around the world but opportunity is not.China has one fifth of the world’s talent and its students want to get into one of the top universities. As a visiting professor I have taught incoming freshmen in Chinese universities who are the equivalent or even better than freshmen in the US. However, when I taught juniors, I noticed a substantial difference in preparation and abilities. This probably happens because of the quality of university education in China. One of the C9 universities invited me to help hire faculty for their CS

Department. After two years, these new hires left the university and I asked them why. They said that the environment was focused on research funding and publication of papers. There was little recognition of quality teaching. There was too much emphasis on seeking research funding and publishing papers.Dedicated faculty at the university that I talked with concurred.

Universities in China need to improve undergraduate education. University presidents have five-year terms and then will be assigned a new job. They believe that if they improve the international ranking of their institution they will be assigned a higher-level position. However, International rankings are based on increasing research funding and the number of papers published. This puts pressure on faculty to spend all their time raising research money and publishing papers in order to be promoted. If presidents desire to raise the ranking of their school, they need to focus on improving the quality of teaching.

Reason to create the program of evaluating undergraduate teaching and recognizing the best faculty

Often the junior faculty work for the senior faculty and do not supervise their own graduate students. Neither fund-raising nor writing papers for publication has anything to do with undergraduate education. It became apparent that somehow the mission of the university, which is to produce the next generation of talent, was being overlooked.

The reason for creating this program of evaluating undergraduate teaching in computer science and recognizing the best faculty was to fix the educational environment in universities and put the focus on quality of teaching.

With the blessing of Premier Li Keqiang who wants quality education for China and the permission of the Ministry of Education, a program was developed to start evaluating the level of teaching. Rating universities on the quality of undergraduate teaching requires a metric. Instead of papers and fundraising,faculty would be judged on the quality of their teaching and their international reputation as viewed by people in the field.The process was funded by a group of companies in China who envisioned this as a long-term investment to increase the talent pool essential to their growth. In 2020, the set of companies include: Bytedance, iFlyTek, Tencent Foundation, and Ping An Technology.

Evaluation considered two issues: the quality of the lecture and the content of the course. To evaluate the quality of the lecture,two emeritus faculty members, or faculty known to be committed to outstanding teaching and who were without bias, sat in on two lectures of each faculty member and used a score card to evaluate the quality of the lecture. The score card consisted of the following items:

Is the faculty member knowledgeable with current material?Is the faculty member excited about the material? Does the faculty member engage the students? What percentage of the students are following the class, at the beginning and at the end? How many are on their cellphones? Did the students arrive on time? Did the faculty member use PowerPoint or the blackboard?

The evaluation process started three years ago when we evaluated teaching in computer science at the top nine universities in China. Two years ago the program was expanded to the top 44 universities in China.

We had planned to recognize these 54 outstanding faculty in Spring of 2020 and had hoped to Start the next round of evaluations of the top 44 institutions by then, but the virus played havoc with our plans.

We had also planned to tell university presidents how well their undergraduate education programs were doing: whether they were in the top quarter, the top half but not the top quarter, or in the bottom half. However, we felt that the presidents were totally focused on running their institution with the virus and that our message would be lost.

We restarted the evaluation this spring and hope to create a ranking of institutions so parents will know which school their children should apply to in order to get the best education. University presidents are of very high quality and talented. They could quickly bring top institutions up to the quality of Stanford, Berkeley, and Cornell.

The mission of universing is to create talent

The mission of a university is to create the next generation of talent, not research. Thus, why do we want faculty engaged in research? The career of a faculty member is roughly 40 years and we want faculty to keep abreast of their evolving field. We do not want them teaching 40-year-old material, if it is no longer important. Curious faculty will keep abreast of the field.Being involved in research is a sign of curiosity. We do not look at how state of the art a faculty member’s research is. A faculty member might just be researching how to find a simpler proof of a known result or a better way to teach some known item.Things we look for in a faculty member is if they are curious and will develop with changes in their field and if they care about the success of all their students.

After World War II the United States realized how important a role science played in the war and made a decision to fund science. They created two classes of science: applied and basic.Applied science is research done to solve a specific problem.The United States has enough scientists and engineers so if a company or the country has a problem, they can hire the researchers necessary to solve the problem. If a company has a problem they need to solve they will hire the talent directly. If the government has a problem they will hire researchers and carry it out in a national laboratory in a timely fashion.

Basic research, on the other hand, is research someone does simply because they are curious about it. There often is no application, only curiosity. A decision was made to carry out basic research in universities where it would train a new generation of researchers. The decision to fund basic research was extremely important. Not only did it educate new researchers but occasionally it led to a new field of research not even envisioned. It created new industries and millions of jobs.

At this moment China has not developed a sufficient quantity of science and engineering talent to meet their need for applied research. Thus, faculty in Chinese universities engage in both applied and basic research. However, if universities do not keep applied and basic research separate, they will never produce the talent that they need. It is appropriate for the university budget to fund basic research. However, applied research should be funded by the companies and government agencies that request it. Ultimately, the applied research should be moved off campus as the US universities like MIT did in creating Lincoln Laboratory and Stanford did in creating Stanford Research Institute.

In the future, education will be important

In the future education will be important. Those nations that create talent in important areas such as AI will attract companies, create new enterprises, and grow their gross domestic product. The importance of nations will be determined by talent rather than physical resources such as coal, oil, or timber.

The advent of AI will automate low-level jobs, which will be eliminated. Higher quality jobs will require a university education. As students get older, they may be in important positions in industry or government and a narrow science education would be a problem. For this reason, it is important that science and engineering curricula include courses in history, the social sciences, and the humanities.

Universities need to educate students not just so they will get an important job but also so that they will enjoy life. During their careers they will have opportunities to enjoy music, sports,travel. They will need exposure to more than just science or engineering.

I found a Chinese Proverb that is perfect for today: If you would plan for one year, plant rice. If you would plan for ten years,plant an apple tree. If you would plan for one hundred years,educate people. Many of our outstanding faculty are planning for a hundred years.They recognize the importance of education and want to make the nation’s environment good for all individuals.