I sympathize a lot with the frustration here and can agree on some of the points (especially the rankings/h-index/etc mania), however...<p>Much of what is bemoaned here is simply a consequence of the fact that science is done by humans and generally costs money, so there is a necessary component of business-type activities (networking, fundraising, marketing, management). Scientists also have human qualities such as egos, desire for money, power, respect and the rest. Why should we be different, or held to a higher standard?<p>Additionally, it's really hard to do original research. First off, really original good ideas are hard to come by; this problem is also not restricted to science or scientists. Then when new ideas come, they are often difficult or impossible to translate into scientific output, because of other constraints, such as lacking good enough data to test hypotheses. And then they typically fail, despite being great ideas. Really good, original, CORRECT ideas are far more rare.<p>Furthermore, science would be much worse off if researchers worked like the author would prefer. If everyone spent 10 years without publishing, working on a completely original problem, there is zero replication. Additionally, if the experiment or approach fails, as they usually do, little is left. So do you publish negative results, and then move on to the next decade's problem?<p>Instead, it's better that the system works the way it does. People do 'bandwagon research', which really is replication. They take the same idea, and apply somewhat different methods, or different datasets, or whatever. And they publish this as often as they can get away with.<p>Then, one can look at the aggregate body of work and evaluate the ideas and methods better. There is more redundancy in the system, and because the papers are smaller, the setbacks are much smaller in the case of failure, and there is a paper trail (so to speak) marking progress along the way. These smaller, more digestible, incremental papers are also more useful to other researchers, who may only be interested in a small part of the work--the methods, or some of the data, or whatever. It would slow everyone down if the only publications were these giant monographs that came out once or twice a decade.<p>It may feel like people are only working on small, unoriginal ideas because many papers look like this, but it's really important to zoom out and see how things fit together. Science is basically a Monte Carlo simulation, and having lots and lots of small and fast iterations covers the space much more efficiently.