Japanese Research is 'No Longer World Class' (nature.com) 35
Japan's contribution to world-class research continues to decline, despite having one of the world's largest research communities, according to a report by the Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT), released in English on 25 October. From a report: Masatsura Igami, the director of the Center for S&T Foresight and Indicators at the National Institute of Science and Technology Policy (NISTEP) in Tokyo, and one of the authors of the 2023 edition of the Japanese Science and Technology Indicators report, says that the findings highlight several areas that Japan could explore to improve its global standing. "Japan's current research environment is far from ideal and is unsustainable. The research environment must shape up," he says.
The report shows that Japan ranks third globally in the total number of researchers, following China and the United States. However, this workforce is not producing the same level of high-impact research as it was two decades ago. Japan's global share of research papers in the top 10% of most-cited articles has slipped from 6% to 2%, intensifying concerns in Japan about its dwindling international standing. Igami explains that the rest of the world has overtaken Japan in terms of quality research output. Some of the decline might be attributable to funding, Igami says. The 2023 report shows that research spending in the university sector has grown by roughly 80% in the United States and Germany, and 40% in France, has quadrupled in South Korea and has increased more than tenfold in China over the past two decades. By contrast, Japan's spending has increased by 10%.
However, even if researchers receive more funding, producing high-impact research might still be challenging, because Japanese scientists have less time for actual research, Igami says. According to a 2020 analysis by MEXT, the proportion of time that university researchers dedicated to science decreased from 47% to 33% between 2002 and 2018. [...] The report's findings confirm those of a previous survey of early-career researchers that pointed to a lack of time for research as a notable factor in job dissatisfaction.
The report shows that Japan ranks third globally in the total number of researchers, following China and the United States. However, this workforce is not producing the same level of high-impact research as it was two decades ago. Japan's global share of research papers in the top 10% of most-cited articles has slipped from 6% to 2%, intensifying concerns in Japan about its dwindling international standing. Igami explains that the rest of the world has overtaken Japan in terms of quality research output. Some of the decline might be attributable to funding, Igami says. The 2023 report shows that research spending in the university sector has grown by roughly 80% in the United States and Germany, and 40% in France, has quadrupled in South Korea and has increased more than tenfold in China over the past two decades. By contrast, Japan's spending has increased by 10%.
However, even if researchers receive more funding, producing high-impact research might still be challenging, because Japanese scientists have less time for actual research, Igami says. According to a 2020 analysis by MEXT, the proportion of time that university researchers dedicated to science decreased from 47% to 33% between 2002 and 2018. [...] The report's findings confirm those of a previous survey of early-career researchers that pointed to a lack of time for research as a notable factor in job dissatisfaction.
Inevitable (Score:5, Insightful)
Declining and aging population, unfriendly immigration, and ridiculous work hours. It's inevitable that Japan's technological status and economy is going to be going downhill.
Re: (Score:1)
> Also, tentacle porn.
That's a sales plus.
lolwut (Score:1)
Japan these days has extremely lenient immigration rules compared to most countries. What other country would let someone from Vietnam move in to work as a convenience store clerk??
What's more of an issue is there is a very finite number of people who speak fluent Japanese who want to move there. Everyone wants to move to the countries that speak English or where there's already a large group of people speaking that language.
That said yes Japan is sliding downhill just like every other developed country, ju
Re: (Score:3)
>What other country would let someone from Vietnam move in to work as a convenience store clerk??
I just assumed most; I know Canada does.
Re: (Score:2)
Japan makes it quite hard to get in for work, so many people doing low level jobs are there on family visas instead. The family visa system is relatively generous.
One issue for people looking to work is that it takes a long time to get Permanent Residency, which is needed for a lot of other important things like being able to get a mortgage. Unless you move when you are fairly young, and keep the goal of PR and settling down in sight, it is hard to make a permanent move.
On top of that wages are relatively l
Re: (Score:2)
Re: (Score:2)
How the hell is quality decided? Often it takes decades to see if ideas pay off. Quantity is easy to count, not quality.
Quality includes both integrity (no cheating), and scientific or technological impact on society.
Re:10 fold increase in China? (Score:5, Informative)
How the hell is quality decided?
Usually by citations. I suppose if you are interested in long-term significance, you could look at citations after five or ten years. A paper that is still getting frequent citations a decade after publishing. For example, according to Google Scholar Stephen A Cook's 1971 paper "The Complexity of Theorem Proving Procedures" was cited over 2200 times in the years between 2017 and 2022. That tells you it is a very high quality paper.
Re: (Score:1)
according to Google Scholar Stephen A Cook's 1971 paper "The Complexity of Theorem Proving Procedures" was cited over 2200 times in the years between 2017 and 2022.
How often did it get sited in the decade after it was published though?
A friend of mine wrote a paper 40 years ago on a medical imaging technique he came up with. It was cited a hand full of times in the first 20 years after being published. It's probably been cited close to 3 or 4 thousand times in the last 20 years.
The problem is that most people aren't that forward thinking. It basically took 20 years for CT and MRI scanners to be able to to use his technique clinically. It's used thousands of times pe
Re: (Score:2)
Re: (Score:2)
In today's culture, I'm not sure this would have happened.
What culture is that? Basis for research often goes very far back to fundamentals. It's not uncommon for modern research to cite papers many decades old.
Re: (Score:1)
> It basically took 20 years for CT and MRI scanners to be able to to use his technique clinically.
The principles of CT and MRI are not that revolutionary in my opinion. It's just a form 3D projection and ray-tracing-like math (laid out in 1971), plus some image enhancing techniques. I agree it takes a lot of experimentation to tune, but any institution with deep pockets could do it.
Computer processing power was probably the real bottleneck.
Re: (Score:1)
> [How is quality decided?] Usually by citations.
A buddy system will cite buddies. I'm sure the Chinese Gov't encouraged citing Chinese studies, even if the fit is borderline.
Re: (Score:2)
Never mind that, where's all the flame wars about having to change the clocks to standard time?
Slashdot is slipping this year.
Two comments (Score:3)
First, there are many ways to measure research contributions: patents, overall papers, papers in "top" conferences, paper citations, university rankings, etc. Unfortunately, none of these metrics capture technology transfer into products. That's the key. There are many product innovations that have come from Japan, but arguably fewer in the last decade. It seems that China and the US are the hot spots for innovation nowadays.
Second, one big problem in Japan (and also Korea and Taiwan, maybe also China) is that young researchers face challenges to explore unconventional ideas and to receive credit for those ideas. I know researchers from East Asia that have moved to the US due to a rigid hierarchical structure that heavily skews freedom and rewards toward the top levels of the hierarchy. The motivation for the grunt researchers to work hard is stunted by this structure, and yet most ideas are generated by the grunt researchers, despite what we might hear in the news.
Re: (Score:3)
Unfortunately, none of these metrics capture technology transfer into products.
That's because it's a stupid metric. It's damn near impossible to measure objectively and the lag between research and practical application can be ridiculous. One study came up with an average lag of 17 years, in that field anyway, but it can span lifetimes.
Re: (Score:2)
Unfortunately, none of these metrics capture technology transfer into products.
That's because it's a stupid metric. It's damn near impossible to measure objectively and the lag between research and practical application can be ridiculous. One study came up with an average lag of 17 years, in that field anyway, but it can span lifetimes.
That's because technology transfer is not a metric but is the desired goal. That's why I lamented that there is no good quantitative metric for tech transfer. However, it's possible to see signs of tech transfer happening. For example, the AI advances of the last few years didn't need 17 years to manifest. Technology maturation time depends on the specific field and technology, and AI advances are happening fast. As a specific example, ChatGPT was released just 5.5 years after the seminal paper from Go
Welcome to the club (Score:1)
It's okay, everyone sucks. Checks and balances are lacking all over.
Relatively Closed Ecosystem (Score:5, Interesting)
Re: (Score:3)
Not completely closed, according to my (dated) experience.
In the 1980's and 1990's Japanese people in computing/AI could read and write English well enough. It was rare for them to speak it or be able to converse in it, though.
Re: Relatively Closed Ecosystem (Score:2)
Re: (Score:2)
The way Japanese is taught at school is focused on passing written exams, so tends to neglect speaking. Japanese people also worry about making mistakes, so speaking is seen as much harder than writing simply because documents can be checked before publication.
In my experience most Japanese engineers have a decent grasp of English too, and often prefer things like datasheets to be in English. When discussing things they will use Japanese though.
Re: (Score:2)
but it seems like everyone speaks English but the Japanese
Not just seems. Japan ranks rather poorly against other rich nations in the EF English Proficiency index. But then their peers they are compared to mostly European or countries where English is the primary language. That stands to reason. Proficiency depends on trade regions. English proficiency is high in Europe because English, French and German and the primary trade languages in Europe. My wife is a teacher here and all three languages are mandatory for her students. In SE Asia English isn't a primary tr
Less time on research? (Score:5, Interesting)
Re: Less time on research? (Score:2)
So where's the failure point? (Score:2)
In China? Due to IP theft and academic deception??
In the US? Due to the decay of actual research and academic deception?
Or an actual degradation in Japan?
How many licks does it take to get to the Tootsie Roll center of a Tootsie Pop?
The world may never know....
skeptical (Score:2)
I don't believe anything in Nature, given all their goofs. Maybe only arcane works in biochemistry I'd believe.
Having worked at a Japanese research company, I can make the following observations:
* Japanese researchers are just as smart, and much less lazy than US researchers.
* Japanese companies are much more secretive than US companies, so a lot less gets published and is kept as proprietary.
* Japanese companies are MUCH more regimented than US companies. That might cut either way.