Lately there has been a discussion going on in Nature about the topic of impact factors.
An Impact factor is more or less how a Journal is 'graded' and is based on how many times articles in that journal are cited by other articles. The logic is of course that the more times a paper is cited the more important it must be.
In recent years it has become mote and more important to publish in high impact factor journals, because money is allocated mainly nowadays based on impact factors. It is an easy way to assess how important a scientists work is.
The problem is that it is not always correct, because not all classic papers in science for instance actually are published in the best journals of that moment.
Also the quest for impact factors leads to rushed publications and splitting up papers.
The question is of course if all of this is in the best interest of science and society?
.... no, of course not
but you have to work under economic constraints, that's just how things are. journals are a profit based industry... you have to make enought money to guarantee your own economic survival. social and ethical obligations may be considered, but it may just be too costly to implement any other system..... *sigh*