Well, that exists in the form of citations. The more a paper is cited the more valued are its authors. Just as the more often a site is linked to, the higher its page rank. And since you're not dealing with massive numbers of sites with lots of SEO experts, straight citation count is good enough without any of the corrections included in the page rank algorithm.

Hiring committees look at both quantity of publications and citations to those publications. The problem is that quantity doesn't always indicate quality. Which is why journal prestige acts as a proxy measure.

I always though this was the impact factor of a journal. However, according to [1] that is not the case. [2] Mentions the 'eigenfactor' which seems to be closer to emulating page-rank. I haven't read much about it though.

[1] https://en.wikipedia.org/wiki/Impact_factor [2] http://researchguides.uic.edu/if/impact

> Well, that exists in the form of citations.

Page-rank is "transitive" in the sense that it not only counts citations, but also citations-of-citations, etc.

I wonder though if such a system can be easily gamed (see SEO), without damaging the reputation of offending authors.

Also, we should perhaps have the concept of "negative citations". For example, when the text contains "In this paper we show that the approach of [..] is incorrect."

Maybe, but it will take some work. IMO the main challenge is to get wide buy in.

Many researchers do not love the current system, but are mostly OK with it. They are not looking to topple it; if everyone is using a different option, they will switch, too, but they are not your passionaries to drive the change.

I agree that good papers are likely the most cited ones. However, how are the people citing a paper going to find it? The traditional mechanism is for the paper to gain exposure by being published in a prestigious journal.