Japan can be a science heavyweight once more if it rethinks funding | Hacker News Reader

Japan can be a science heavyweight once more if it rethinks funding

nature.com

159 pointsrntn51 comments

This article seems interested in suggestion Japan abandon its unique approach and adopt the approach used by other nations. That is silly.

Japan is one of the foremost funders of deep research. It funds large physics experiments. It has a long history of semiconductor innovations. MEXT scholarships have proven a brilliant method to attract smart men and women from around the world.

Here in Touhoku I've met so many bright international students on MEXT scholarships doing research within those exact project-funded teams. Switching to person-focused funding would be silly, do you really think a smart guy from Congo is going to be able to win funding? That Japan has a system where the product/team can focus on an established topic, then backfill with smart researches, is a strength not a weakness.

Of course Nature is in the business of publishing papers, not science. So it makes sense they would be blind to the reality of science: you measure it in results not papers. The academics I know are all focused on achieve specific goals, they rarely talk about the papers in the way the Canadian Acedemics I know did back home. Think "I want to automate boar trap monitoring so that farmers do not need to check it everytime, and so that non-boars do not get trapped". That is the sort of highly practical research you get when a supervisor knows their field and knows their country. It might not pay off in papers, but it will pay off for Japan as a country.

The world should be taking lessons from Japan, not the other way around. Team based funding. Scholarships for bright students from any country. Deep funding for physical research other than just ITER and LHC.

That's interesting.

Japanese discussions I have had, and articles in Japanese seem to say the opposite. That Japan needs to invest more; because many talented Japanese researchers are emigrating to the USA or China.

The main topic that comes up is that both China and the USA provide better wages, as well as greater funding for projects overall.

But I'm just a local, and certainly not a researcher. Thanks for your POV!

To be honest, those are all valid points which I think are true. I've never met an academic who thinks their country should reduce science funding. And the personal incentives do push researchers overseas for higher wages.

Personally though, I think how a country uses the money dominates over how much. Most countries have a pretty consistent level of funding. Sure some countries might double others, but overall funding tends to follow GDP. No country is spending 10%+ of GDP on research, nor do I think is that justifiable.

Thus the differences come from effectiveness of spend, not volume. Japan has an advantage here in the low English proficiency: you cannot be headhunted by the Americans if you cannot speak English. Thus when Japan does focus on specializations, as it did in the past with semiconductors, those researchers cannot be headhunted away.

Are talented Japanese researchers emigrating to China? Not that I don't think there's good work going on there, but my impression from the outside is it's extremely competitive.

I definitely knew a lot of people who end up heading over to the US though. Lot of people who come back because living in the US is miserable for a lot of em though! The salary gap is huge (especially with the exchange rate) but at the end of the day Tokyo in particular has a lot going for it.

I really like Japan uni's being fairly researched focused, though. It's to the point where universities with exchanges with Japanese universities were complaining because people in their home countries had credit requirements for Masters students but Japanese unis were like "I mean you're a Masters student, go do research and take a couple classes. But mainly do research". At least that's what I was told.

But at the end of the day research labs in Japan can't escape general work culture pressures of the country. A certain very famous prolific architect gets all his work done by farming it out to his masters students at the university. Professors in many labs basically only live and breath the academic life, so graduate students are just constantly being asked to be in "work mode". Just no break whatsoever. Hell of a lot of churning yet going nowhere. Might be the case in academia elsewhere but I feel like people at least get a bit more vacation time.

All of that, and you're paying for the privilege as a masters student!

> I definitely knew a lot of people who end up heading over to the US though. Lot of people who come back because living in the US is miserable for a lot of em though!

As a country, Japan (probably) wants the opposite, where the talented spend their most productive years in Japan and the most negative-productive (elderly healthcare costs, pension, etc.) years elsewhere. I'm not denigrating the individuals though, ask not what you can do for your country, but what your country can do for you.

Universities in Europe have bachelor's be 3 years and Universities in the US are 4 years but students need to take a lot of non-major-related classes. In Europe it is almost a given you will be doing a 2 year masters after bachelor if you are going to any STEM degree.

In Brazil bachelor is usually 4 years (sometimes 5) and it is hardcore STEM all the way through with higher workload than EU Unis per semester. Masters in Brazil are like you said, you might take ~12-16 credits over 2 years, you are actually expected to be doing research and writing your thesis the rest of the time. From what I talked with friends it is like that in India and China as well (maybe Japan too?).

I don't have a masters myself, but I hear masters in Brazil are pretty chill, while in the EU it can be quite demanding. In EU you need to write a thesis and handle the class workload that is not minimum like it is in Brazil during the masters. But Bachelor in Brazil is waaaay more intense and lasts longer too.

It is quite annoying when sending CVs in Europe people see I only have a bachelor and I think I don't know deeper Computer Science/Math concepts.

England and Netherlands are 3 years for a BSc but that’s not the case for all other European countries. Even in Scotland it’s 4 years for a BSc.

Even in the U.K. they realised 20 years ago that that wasn’t enough so they introduced ´integrated masters’ as a preparation for additional study, where you spend another year and come out with a single MSci or MEng degree instead of a BSc at the end of it.

In Sweden almost everyone I meet while working have a masters while in Brazil it is the opposite. Very few people go for masters in Brazil, the companies simply don't care about it at all so it is mostly just people who want to get PhDs afterwards.

Funny thing is, most people in Brazil who do Masters usually get an stipend too and it is much more chill than the bachelor. Yet most still go to private sector because the money is better.

In Sweden people actively look down on your CV if you only have a bachelor.

Probably to note that in Asia, professors are generally expected to retire at 60, with only very rare exceptions.

I don't think that's fundamentally different from the US model.

While some countries have general funding calls where individual PIs can request funding for basically anything, the US model is based on large decentralized projects. Some academics spend a few years at funding agencies, identifying topics that could benefit from focused research. Then the agency issues a funding call, inviting applications from PIs who believe they can contribute. With some agencies such as NSF, this structure is more nominal, as individual grants are usually too small to hire permanent staff. But others such as NIH award larger grants, which make hiring professional researchers and support staff possible.

The fundamental issue with goal-oriented research is that it will narrow your vision. Topics don't get funded if there are no reasonable expectations that the idea will work and be beneficial. But that leaves the academia in an awkward position. You want topics that are promising enough to get funded but not so promising that the industry will also pursue them with better funding and higher salaries.

Goal-oriented research needs to be balanced with curiosity-driven research. Research focused on things the researchers find interesting, without any expectation that it will be beneficial to someone. Most of the time it won't be beneficial, but occasionally you get unexpected breakthroughs.

We wouldn't have blue leds if the Japanese used the American model

I doubt you can attribute Shuji Nakamura's success entirely to the "Japanese model".

https://www.youtube.com/watch?v=AF8d72mA41M

The man was a genius, and his success is mostly his alone.

Right, but also left alone to do what he wanted for years without revenue is not the American model.

I don't think that's the Japanese model either, nor is it a good one.

Lots of grad-students are "abandoned" with funding and, unless they're extremely extroverted & good at networking, always fail badly.

I'd say it's implicitly a result of the Japanese work model, not academic model.

Employed until retirement kind of hiring was the norm in Japan, with the work over life focus meaning many companies saw their peers more as an tight knit fraternity rather than the dry "strictly business" hiring in west.

So someone with known track record of being diligent and hardworking was often given more chances, even if his work didn't immediately produce result.

You're right that it's not a good one in the sense that for majority of the cases this just leads to bloat and unproductive members sticking around. But many innovations, academic or industry, in Japan, came from people being measurably bad/average in productivity, given a chance until they found their breakthrough. (entertainment industry had a lot of those)

Very interesting; that makes a lot more sense (and definitely worth emulating).

How much of the US structure is still in place and unaffected by EOs?

It's true tho that Japanese universities have a problem with labs communication. In the former university I was a MEXT student in, there was at least three labs that could have collaborated on topics that my lab was focused one. Granted, one professor welcomed me in his lab to do some stuffs with his students. But otherwise it was like little fiefdoms taking pride in doing things alone. This is contrasting with what's going on (at least in intention) in French research.

>Japan is one of the foremost funders of deep research.

Perhaps they were once upon a time, it's questionable if they still are today.

And anyway, that's not even the real problem Japan's R&D world has: The real problem is that Japan can't bring products to market. If you can't make some money your R&D is going to eventually run out of gas and die.

Japan is also poorer than its financials might suggest. Japanese society shuns pioneering, the road not yet taken might as well be the plague; everyone wants to be #2 or lower, not #1. This means a lot of Japanese capital ends up sitting around doing nothing besides accruing petty interest, only a small fraction gets budgeted for pioneering into the new and the unknown after significant disparagement.

>The academics I know are all focused on achieve specific goals, they rarely talk about the papers in the way the Canadian Acedemics I know did back home. Think "I want to automate boar trap monitoring so that farmers do not need to check it everytime, and so that non-boars do not get trapped".

I guess one of the bigger risks of doing science the Japanese way would be the development of pet projects and tunnel vision leading to dead-ends. This is by no means limited to Japan, but the number of cases where this has happened (hydrogen fuel cells, Honda robots, TRON OS) are notable.

Main topic aside, what's your position on research in Tsukuba? I've read several articles suggesting it's become a waste of money, infrastructure, and manpower as a number of labs (aside from JAXA) and private research firms (mainly in semi conductors) have either shuttered or moved elsewhere (mainly Tokyo). While Tsukuba may be a bit out of your way in both location and relevant subject matter, I would appreciate any insight you might have.

> Of course Nature is in the business of publishing papers, not science

The irony of the statement perhaps can be extended, Science (journal) is in the business of publishing papers, not science

Japan used to be known as the one nation in the world where old people were fans of bran new tech.

However, today it would seem that's because that generation lived through the great changes Japan experienced. Extremely quickly going from a pre-industrial civilization, to a post-industrial one.

And so despite Japan's great traditionally intense conservatism, they were fans of technological innovation.

With that generation fading away, it seems Japan is returning to being hyper conservative in every way. And falling behind technologically.

The story of Shuji Nakamura, inventor of the blue LED, is interesting. The company founder supported his experimentation, but the next generation, the son of the founder, wanted to Shuji Nakamura to stop.

Faith in innovation vs pessimism.

I don't follow what this article is trying to say - if the point is to have more interdisciplinary research, isn't project-focused funding precisely what's needed? Researcher focus seems like it would focus on that researcher's discipline, not an interdisciplinary team.

In fact, my experience in a Japanese lab made me feel funding precisely goes towards researchers rather than projects if professors count as researchers. Professors get funding for their labs, and while there will be project proposals the perceived value of that professor is key for closing it.

Maybe the article is focused more on private research labs than universities though.

Nobody can be a real scientific heavy weight anymore, whatever filters against that is in the culture ,in the institutions. we filter for knowledge reproduction automation .

[stub for offtopicness]

[see https://news.ycombinator.com/item?id=43018086 for why]

It can probably become a science heavyweight simply by being here at the moment when science funding in the USA is about to collapse.

The USA did quite well in applied research before the federal government became the dominant source of funds. That said, it would probably take some time (and some pain) to readjust, and theoretical/arts research would likely be dramatically reduced.

> The USA did quite well in applied research before the federal government became the dominant source of funds.

For my own curiosity, when did that transition occur?

1950 according to wikipedia.

https://en.wikipedia.org/wiki/National_Science_Foundation

DARPA, ARPA-H, etc.

Can you share any pointers on this topic about applied research success, I'm guessing in the pre-WW2 era? I have not heard anything about that and have not been able to locate supporting resources by a web search just now.

Edison, Bell, IBM, Morse, US Steel, Wright, Sikorsky, Westinghouse, Marconi, Dictaphone, Goodyear...there's a lot, any many of the companies are still household names even if they don't exist anymore.

How did these R&D operations get their start? The same as universities at the time: they were funded by private backers, often family, or the parent corporation of the R&D Lab (IBM Research, Bell Labs, General Electric, etc.).

It's Vannevar Bush who spearheaded the creation of the federal research funding system as we know it today, for the sake of the war effort: https://en.wikipedia.org/wiki/Vannevar_Bush.

As a side effect, basic research is now mostly government funded, and the peer review system was created to ensure taxpayers were "getting their money's worth".

A more interesting question is: why did it work then, and could it work today? My (admittedly pessimistic) view is that life-changing innovations were a lot easier (read: cheaper) to create back then. Continual breakthroughs in materials science, transportation and communications technologies left a lot of "white space" to innovate in.

Perhaps US Universities should tap into billion dollar endowments or slash the administrative headcount instead of raising tuition yet again.

https://www.theatlantic.com/health/archive/2025/02/nih-trump...

But the point of an endowment is to have a relatively stable funding stream in perpetuity - if you start dipping into the principal, the funding stream starts shrinking and is no longer perpetual.

Harvard’s endowment, for example, already funds a little less than 40% of its budget, and should be able to continue doing so indefinitely. If you bump that to even, say, 50% there’s a good chance it won’t have any endowment (or funding stream!) at all in the year 2100.

Arguably, the USA did even better with the feds funding research. I don't know of any program that had a higher ROI than DARPA's VLSI Project.

https://en.wikipedia.org/wiki/VLSI_Project

> The VLSI Project is one of the most influential research projects in modern computer history. Its offspring include Berkeley Software Distribution (BSD) Unix, the reduced instruction set computer (RISC) processor concept, many computer-aided design (CAD) tools still in use today, 32-bit graphics workstations, fabless manufacturing and design houses, and its own semiconductor fabrication plant (fab), MOSIS, starting in 1981.[2] A similar DARPA project partnering with industry, VHSIC had little or no impact.

Wow that's insane, I never knew that

Please don't take HN threads on generic tangents. (This is in the site guidelines: https://news.ycombinator.com/newsguidelines.html.)

It's clear why this happens and of course it's not ill intended. The mind naturally follows the most-traveled association path from a new stimulus back to something familiar.

Unfortunately, that is the anticurious direction. Swapping out a specific new topic for the nearest familiar one means replacing a potentially new and interesting discussion with a repetitive old one.

These large generic themes are like black holes: if you fly too close to one, you get sucked in there instead of going wherever else you might have explored, and then no new information emerges.

https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...

I do get your point, and generally think it makes sense, but in this case it seems extremely relevant: There is a global competition for intellectual capital, and at the moment, the US' position here seems at the very least uncertain, given all that is happening.

> I do get your point, [...] but in this case it seems extremely relevant

Not sure you do.

At a minimum, they could rephrase for better discussion v. a snarky, low effort claim that the sky is falling.

The sky is indeed falling. You not wanting to look up doesn't change that.

You're still not taking a hint [0].

[0] https://news.ycombinator.com/item?id=43018477

dang knows HN better than anyone. I was coming here to discuss it and don't remember seeing much of it. I don't know how you moderate a forum, in this sense, where each person sees only a small segment of it. It will be overdone to some and novel to others.

(Some things are probably overdone for everyone.)

> each person sees only a small segment of it. It will be overdone to some and novel to others.

You've said concisely what I just spent an hour saying verbosely: https://news.ycombinator.com/item?id=43019507.

The US is literally cutting funding now for science and letting politics get in the way of science.

We can’t have an honest discussion about this without addressing the elephant in the room.

Japan or some other nation has a chance to step up and fill the void that the US is creating. Some other countries universities could even partner with the US universities.

Sure, but there have been many threads about that*, there can and will be others, and those threads are the place to discuss it.

Not allowing large/important/hot/generic/divisive topics to drown out smaller/quieter/marginal/curious/specific ones is one of the core principles here.

* Edit: here's a partial list:

What's happening inside the NIH and NSF - https://news.ycombinator.com/item?id=42940257 - Feb 2025 (1519 comments)

CDC: Unpublished manuscripts mentioning certain topics must be pulled or revised - https://news.ycombinator.com/item?id=42905937 - Feb 2025 (719 comments)

CDC data are disappearing - https://news.ycombinator.com/item?id=42897696 - Feb 2025 (589 comments)

NSF starts vetting all grants to comply with executive orders - https://news.ycombinator.com/item?id=42886661 - Jan 2025 (488 comments)

'Never seen anything like this' – NIH meetings and travel halted abruptly - https://news.ycombinator.com/item?id=42817910 - Jan 2025 (111 comments)

NIH hit with freezes on meetings, travel, communications, and hiring - https://news.ycombinator.com/item?id=42798960 - Jan 2025 (440 comments)

nah. It's not funding. It's more a side effect of japans economic problems.

At this point is the US that needs to rethink funding if it wants to continue to be competitive. The US funding for universities has been frozen (adjusted for inflation) for the last 20 years. It was enough when the rest of the world invested much less in science. But nowadays China is moving faster while US funding is still the same. Giving money to tech companies is not a substitute, because they only care about short term gains.

We can also improve the efficiency of allocation of existing funds.

It's ridiculous that returns on public basic research have been so great and yet funding has been so neglected. The chips act was a small positive step.

This was the biggest change I expected from Biden, however it seems that giving hundreds of billions to big companies is the priority, instead of supporting national research institutions.

I want to point out that money is not the only reason why Samsung and TSMC accept this idea. Political reasons should play an important role here, too.

Reply on news.ycombinator.com