Hacker News Reader

Comment by simonsarris | Hacker News Reader

> I generally disagree with the premise that scientists are becoming less productive.

Do you believe the average academic research paper or project written/done today is of the same quality or better as the average in 1980, or 1950?

View on HN

I can't speak to all science, but I am a CS (cryptography) researcher and routinely read papers from the 1980s. The level of rigor and quality of my field's papers has absolutely improved, by leaps and bounds. The formal definitions in those early papers are often non-existent (and sometimes wrong in retrospect) and the proposed constructions are often much simpler (and sometimes subtly wrong in retrospect.) And the number of papers has increased by at least an order of magnitude.

On the flip side, those early papers contain the most fundamental discoveries in our field: you're only going to invent RSA or blind signatures or zero knowledge once. It's possible all those researchers were much smarter than we are now. (I grant this!) But there are a lot of absolutely brilliant people I know today. Alternatively, the lower-hanging fruit is all gone and the problems have become much harder.

> Alternatively, the lower-hanging fruit is all gone and the problems have become much harder.

Reminds me of a patent officer who said in 1899 that "everything that can be invented has been invented."

What's amusing about my field is that while much of the rigor was lost (computational biology in the 90s was very CS-rigorous), what we've learned is that deep networks beat any human features, none of the rigors of chomsky hierarchy really matter to find interesting biology, and you don't even need to know how to differentiate because that's automatic now.

Agreed, all those formally simulated cells and metabolic networks didn't age well.

> you're only going to invent RSA or blind signatures or zero knowledge once.

Well, maybe. Sometimes things that are discovered are lost or go completely unnoticed, so progress is not always monotonic. For example, for a long time it was believed that John William Strutt first documented dynamic soaring [1] in birds in the 1800s, but just recently (2018) it was found that actually none other than Leonardo DaVinci documented the phenomena in his notebooks centuries before [2]. People just didn't notice or forgot, despite those documents being some of the most poured over in history.

[1] https://en.wikipedia.org/wiki/Dynamic_soaring

[2] https://royalsocietypublishing.org/doi/10.1098/rsnr.2018.002...

Sometimes people just aren't ready for new knowledge or they don't know enough to understand it. I see this when new person enters an industry, healthcare in my case. It takes about a year to understand the nuances between healthcare providers, funders and administrators. I am guessing people read what Leonardo wrote but they just didn't understand the significance.

Good point. The same thing definitely happens in physics too. For example, people attributed important advances in mathematical physics (such as the mean speed theorem) to Galileo despite them having been already developed extensively in the middle ages.

Technically - yes. But it 's just that, a paper. Back then , people were chasing glory, not "the paper" because papers didnt matter, that s why so many of them were too short, unreferenced etc. Science was small and more personal. Now it's an impersonal industry and science is a product. Science gets a lot of underserved respect out of inertia, but it really is a big industry sector now.

I can only talk about computer science. The research papers in the past would probably not be able to get published in prestigious journals today. They'd need to get expanded from sketchy 8 page papers into very rigorous 25 page papers. I think we've overshot though. With papers in the modern style you need to wade through a lot of cruft to get to the key idea. As for the quality of the ideas...the older papers do win. I think that's because there was more low hanging fruit, but some part of it may be due to an incentive structure that rewards safe bets.

> quality

What do you mean by that? Good writing? Good rigor? Good research work? Have a fundamental discovery? There are not as correlated as one expects.

This is a trick question. How do you quantify quality of articles? How much new science they created? We can't know that yet. What they provided in term of societal impact? SameThey reproductibility, lot of chemistry and biology from the 50s turned out to be really wrong for example. I would still say that the average quality (in my field), chemistry is lower but just because there is a deluge of really low quality publishers and journals that have no ethics. But if you stay in the major journals of fields, the answer is radically different.

In terms of writing and quality of research methodology, far higher in my experience

1950 did not even had peer review. Peer review as we know it started in 1970.

Reply on news.ycombinator.com