Hacker News Reader

> if it is true that 50% of all papers written never get an audience

This can not be assessed. Even if you could somehow determine whether a paper was read or not, how long after publication do you wait until you declare it "never read"?

View on HN

If it's not read in even a year or two, I think that can still be considered a failure.

That is far too shortsighted.

Yeah, especially when you consider all the current buzz about Neural Networks. The original papers on MLPs, CNNs, etc were published more than two decades ago but were not given much attention until recently.

And I certainly wouldn't consider that a failure.

I'm sure there are other examples of research catching on decades later.

Many of the most important advances in science are ignored for decades. To take one well known example, Mendel’s work on breeding of pea plants was published in the 1860s in an obscure journal and completely ignored by biologists at the time, but was rediscovered and popularized 40+ years later and now forms the foundation of the modern understanding of genetics.

Reply on news.ycombinator.com